diff --git "a/slurm.out" "b/slurm.out"
new file mode 100644--- /dev/null
+++ "b/slurm.out"
@@ -0,0 +1,3023 @@
+1: [2025-08-06 23:16:07,347] [INFO] [axolotl.utils.schemas.validation.check_eval_packing:118] [PID:2862582] [RANK:0] explicitly setting `eval_sample_packing` to match `sample_packing`
+3: [2025-08-06 23:16:07,347] [INFO] [axolotl.utils.schemas.validation.check_eval_packing:118] [PID:391404] [RANK:0] explicitly setting `eval_sample_packing` to match `sample_packing`
+2: [2025-08-06 23:16:07,347] [INFO] [axolotl.utils.schemas.validation.check_eval_packing:118] [PID:1823156] [RANK:0] explicitly setting `eval_sample_packing` to match `sample_packing`
+0: [2025-08-06 23:16:07,349] [INFO] [axolotl.utils.schemas.validation.check_eval_packing:118] [PID:411490] [RANK:0] explicitly setting `eval_sample_packing` to match `sample_packing`
+1: [2025-08-06 23:16:07,347] [INFO] [axolotl.utils.schemas.validation.hint_sample_packing_padding:217] [PID:2862582] [RANK:0] Setting `pad_to_sequence_len: true` to prevent memory leaks when sample_packing
+3: [2025-08-06 23:16:07,347] [INFO] [axolotl.utils.schemas.validation.hint_sample_packing_padding:217] [PID:391404] [RANK:0] Setting `pad_to_sequence_len: true` to prevent memory leaks when sample_packing
+2: [2025-08-06 23:16:07,348] [INFO] [axolotl.utils.schemas.validation.hint_sample_packing_padding:217] [PID:1823156] [RANK:0] Setting `pad_to_sequence_len: true` to prevent memory leaks when sample_packing
+0: [2025-08-06 23:16:07,349] [INFO] [axolotl.utils.schemas.validation.hint_sample_packing_padding:217] [PID:411490] [RANK:0] Setting `pad_to_sequence_len: true` to prevent memory leaks when sample_packing
+0: [2025-08-06 23:16:16,186] [INFO] [axolotl.cli.config.load_cfg:244] [PID:411490] [RANK:0] config:
+0: {
+0:   "activation_offloading": false,
+0:   "auto_resume_from_checkpoints": true,
+0:   "axolotl_config_path": "/lustre/fswork/projects/rech/dgo/udv55np/train/tmp/1754476677918612596.yaml",
+0:   "base_model": "/lustre/fswork/projects/rech/qwv/udv55np/Qwen/Qwen2.5-0.5B",
+0:   "base_model_config": "/lustre/fswork/projects/rech/qwv/udv55np/Qwen/Qwen2.5-0.5B",
+0:   "batch_size": 16,
+0:   "bf16": true,
+0:   "capabilities": {
+0:     "bf16": true,
+0:     "compute_capability": "sm_90",
+0:     "fp8": false,
+0:     "n_gpu": 4,
+0:     "n_node": 1
+0:   },
+0:   "chat_template": "qwen_25",
+0:   "dataloader_num_workers": 4,
+0:   "dataloader_pin_memory": true,
+0:   "dataloader_prefetch_factor": 256,
+0:   "dataset_prepared_path": "/lustre/fsn1/projects/rech/dgo/udv55np/dataset/Qwen3-235B-A22B/Qwen2.5-0.5B/1",
+0:   "dataset_processes": 192,
+0:   "datasets": [
+0:     {
+0:       "chat_template": "tokenizer_default",
+0:       "data_files": [
+0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0007.jsonl",
+0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0009.jsonl",
+0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0005.jsonl",
+0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0006.jsonl",
+0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0014.jsonl",
+0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0010.jsonl",
+0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0012.jsonl",
+0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0008.jsonl",
+0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0001.jsonl",
+0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0002.jsonl",
+0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0013.jsonl",
+0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0015.jsonl",
+0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0004.jsonl",
+0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0011.jsonl",
+0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0000.jsonl",
+0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0003.jsonl"
+0:       ],
+0:       "ds_type": "json",
+0:       "field_messages": "conversations",
+0:       "message_property_mappings": {
+0:         "content": "content",
+0:         "role": "role"
+0:       },
+0:       "path": "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking",
+0:       "trust_remote_code": false,
+0:       "type": "chat_template"
+0:     }
+0:   ],
+0:   "ddp": true,
+0:   "deepspeed": {
+0:     "bf16": {
+0:       "enabled": true
+0:     },
+0:     "gradient_accumulation_steps": "auto",
+0:     "gradient_clipping": "auto",
+0:     "train_batch_size": "auto",
+0:     "train_micro_batch_size_per_gpu": "auto",
+0:     "wall_clock_breakdown": false,
+0:     "zero_optimization": {
+0:       "contiguous_gradients": true,
+0:       "overlap_comm": true,
+0:       "reduce_bucket_size": "auto",
+0:       "stage": 3,
+0:       "stage3_gather_16bit_weights_on_model_save": true,
+0:       "stage3_param_persistence_threshold": "auto",
+0:       "stage3_prefetch_bucket_size": "auto",
+0:       "sub_group_size": 0
+0:     }
+0:   },
+0:   "device": "cuda:0",
+0:   "device_map": {
+0:     "": 0
+0:   },
+0:   "env_capabilities": {
+0:     "torch_version": "2.6.0"
+0:   },
+0:   "eval_batch_size": 1,
+0:   "eval_causal_lm_metrics": [
+0:     "sacrebleu",
+0:     "comet",
+0:     "ter",
+0:     "chrf"
+0:   ],
+0:   "eval_max_new_tokens": 128,
+0:   "eval_sample_packing": true,
+0:   "eval_table_size": 0,
+0:   "evals_per_epoch": 0,
+0:   "flash_attention": true,
+0:   "fp16": false,
+0:   "gradient_accumulation_steps": 4,
+0:   "gradient_checkpointing": true,
+0:   "gradient_checkpointing_kwargs": {
+0:     "use_reentrant": true
+0:   },
+0:   "learning_rate": 2e-05,
+0:   "lisa_layers_attribute": "model.layers",
+0:   "load_best_model_at_end": false,
+0:   "load_in_4bit": false,
+0:   "load_in_8bit": false,
+0:   "local_rank": 0,
+0:   "logging_steps": 10,
+0:   "lora_dropout": 0.0,
+0:   "loraplus_lr_embedding": 1e-06,
+0:   "lr_scheduler": "warmup_stable_decay",
+0:   "lr_scheduler_kwargs": {
+0:     "min_lr_ratio": 0.1,
+0:     "num_decay_steps": 300
+0:   },
+0:   "max_prompt_len": 512,
+0:   "mean_resizing_embeddings": false,
+0:   "micro_batch_size": 1,
+0:   "model_config_type": "qwen2",
+0:   "num_epochs": 1.0,
+0:   "optimizer": "adamw_torch_fused",
+0:   "output_dir": "/lustre/fswork/projects/rech/dgo/udv55np/ift/Qwen3-235B-A22B/Qwen2.5-0.5B/1",
+0:   "pad_to_sequence_len": true,
+0:   "pretrain_multipack_attn": true,
+0:   "pretrain_multipack_buffer_size": 10000,
+0:   "profiler_steps_start": 0,
+0:   "qlora_sharded_model_loading": false,
+0:   "ray_num_workers": 1,
+0:   "resources_per_worker": {
+0:     "GPU": 1
+0:   },
+0:   "sample_packing": true,
+0:   "sample_packing_bin_size": 200,
+0:   "sample_packing_group_size": 100000,
+0:   "save_only_model": false,
+0:   "save_safetensors": true,
+0:   "save_steps": 0.2,
+0:   "save_total_limit": 20,
+0:   "sequence_len": 16384,
+0:   "sequence_parallel_degree": 1,
+0:   "shuffle_merged_datasets": true,
+0:   "skip_prepare_dataset": false,
+0:   "special_tokens": {
+0:     "bos_token": "<|im_start|>",
+0:     "eos_token": "<|im_end|>",
+0:     "pad_token": "<|endoftext|>"
+0:   },
+0:   "strict": false,
+0:   "tensor_parallel_size": 1,
+0:   "tf32": false,
+0:   "tiled_mlp_use_original_mlp": true,
+0:   "tokenizer_config": "/lustre/fswork/projects/rech/qwv/udv55np/Qwen/Qwen2.5-0.5B",
+0:   "torch_dtype": "torch.bfloat16",
+0:   "train_on_inputs": false,
+0:   "trl": {
+0:     "log_completions": false,
+0:     "mask_truncated_completions": false,
+0:     "ref_model_mixup_alpha": 0.9,
+0:     "ref_model_sync_steps": 64,
+0:     "scale_rewards": true,
+0:     "sync_ref_model": false,
+0:     "use_vllm": false,
+0:     "vllm_server_host": "0.0.0.0",
+0:     "vllm_server_port": 8000
+0:   },
+0:   "use_ray": false,
+0:   "use_tensorboard": true,
+0:   "val_set_size": 0.0,
+0:   "vllm": {
+0:     "device": "auto",
+0:     "dtype": "auto",
+0:     "gpu_memory_utilization": 0.9,
+0:     "host": "0.0.0.0",
+0:     "port": 8000
+0:   },
+0:   "warmup_steps": 150,
+0:   "weight_decay": 0.0,
+0:   "world_size": 4
+0: }
+0: [2025-08-06 23:16:16,187] [INFO] [axolotl.cli.checks.check_user_token:35] [PID:411490] [RANK:0] Skipping HuggingFace token verification because HF_HUB_OFFLINE is set to True. Only local files will be used.
+1: [2025-08-06 23:16:16,465] [INFO] [axolotl.utils.data.sft._load_raw_datasets:310] [PID:2862582] [RANK:0] Loading raw datasets...
+1: 
Downloading data:   0%|          | 0/16 [00:00<?, ?files/s]
Downloading data: 100%|██████████| 16/16 [00:00<00:00, 31610.39files/s]
+1: 
Generating train split: 0 examples [00:00, ? examples/s]
Generating train split: 1002 examples [00:00, 6770.97 examples/s]
Generating train split: 1930 examples [00:00, 7227.93 examples/s]
Generating train split: 7674 examples [00:00, 23035.52 examples/s]
Generating train split: 15584 examples [00:00, 39710.54 examples/s]
Generating train split: 23318 examples [00:00, 49558.20 examples/s]
Generating train split: 31067 examples [00:00, 56011.87 examples/s]
Generating train split: 38856 examples [00:00, 60297.65 examples/s]
Generating train split: 46493 examples [00:00, 62357.20 examples/s]
Generating train split: 54316 examples [00:01, 63894.04 examples/s]
Generating train split: 62208 examples [00:01, 65238.00 examples/s]
Generating train split: 70112 examples [00:01, 67004.53 examples/s]
Generating train split: 77845 examples [00:01, 67996.00 examples/s]
Generating train split: 85692 examples [00:01, 68901.07 examples/s]
Generating train split: 93740 examples [00:01, 39634.04 examples/s]
Generating train sp
+1: lit: 100612 examples [00:02, 43805.06 examples/s]
Generating train split: 108408 examples [00:02, 49582.31 examples/s]
Generating train split: 116183 examples [00:02, 54345.87 examples/s]
Generating train split: 124004 examples [00:02, 58349.91 examples/s]
Generating train split: 131832 examples [00:02, 61429.94 examples/s]
Generating train split: 139607 examples [00:02, 63706.06 examples/s]
Generating train split: 147620 examples [00:02, 65213.36 examples/s]
Generating train split: 155401 examples [00:02, 66918.18 examples/s]
Generating train split: 163343 examples [00:02, 69060.62 examples/s]
Generating train split: 171069 examples [00:03, 70205.49 examples/s]
Generating train split: 178942 examples [00:03, 53199.41 examples/s]
Generating train split: 185657 examples [00:03, 53738.63 examples/s]
Generating train split: 193506 examples [00:03, 58506.50 examples/s]
Generating train split: 201318 examples [00:03, 61228.83 examples/s]
Generating train split: 209030 examples [00:03, 64082.29 examples/s]
Generati
+1: ng train split: 216747 examples [00:03, 66112.28 examples/s]
Generating train split: 224597 examples [00:03, 67444.23 examples/s]
Generating train split: 232511 examples [00:04, 68826.51 examples/s]
Generating train split: 240342 examples [00:04, 69906.64 examples/s]
Generating train split: 248121 examples [00:04, 69385.45 examples/s]
Generating train split: 255863 examples [00:04, 70633.56 examples/s]
Generating train split: 263347 examples [00:04, 53823.53 examples/s]
Generating train split: 271072 examples [00:04, 51115.34 examples/s]
Generating train split: 278839 examples [00:04, 55704.59 examples/s]
Generating train split: 286659 examples [00:04, 60052.92 examples/s]
Generating train split: 294361 examples [00:05, 62357.29 examples/s]
Generating train split: 302180 examples [00:05, 65109.64 examples/s]
Generating train split: 309905 examples [00:05, 66673.21 examples/s]
Generating train split: 317545 examples [00:05, 67410.54 examples/s]
Generating train split: 325270 examples [00:05, 68327.06 examples/
+1: s]
Generating train split: 332969 examples [00:05, 69437.48 examples/s]
Generating train split: 340668 examples [00:05, 68709.77 examples/s]
Generating train split: 349406 examples [00:05, 58531.07 examples/s]
Generating train split: 357240 examples [00:06, 45747.81 examples/s]
Generating train split: 365042 examples [00:06, 51195.60 examples/s]
Generating train split: 372775 examples [00:06, 55624.48 examples/s]
Generating train split: 380749 examples [00:06, 59741.29 examples/s]
Generating train split: 388476 examples [00:06, 61426.14 examples/s]
Generating train split: 396327 examples [00:06, 64255.32 examples/s]
Generating train split: 404035 examples [00:06, 65869.66 examples/s]
Generating train split: 411851 examples [00:06, 67739.09 examples/s]
Generating train split: 419618 examples [00:07, 69427.31 examples/s]
Generating train split: 427562 examples [00:07, 71128.76 examples/s]
Generating train split: 435298 examples [00:07, 71463.83 examples/s]
Generating train split: 445212 examples [00:07, 35615.7
+1: 7 examples/s]
Generating train split: 452962 examples [00:07, 41252.89 examples/s]
Generating train split: 460725 examples [00:08, 46769.35 examples/s]
Generating train split: 468517 examples [00:08, 51612.00 examples/s]
Generating train split: 476378 examples [00:08, 55994.51 examples/s]
Generating train split: 484128 examples [00:08, 58688.61 examples/s]
Generating train split: 491968 examples [00:08, 61993.90 examples/s]
Generating train split: 499820 examples [00:08, 64553.45 examples/s]
Generating train split: 507574 examples [00:08, 66797.30 examples/s]
Generating train split: 515375 examples [00:08, 68836.95 examples/s]
Generating train split: 523674 examples [00:08, 62324.37 examples/s]
Generating train split: 532563 examples [00:09, 41583.37 examples/s]
Generating train split: 540513 examples [00:09, 47676.58 examples/s]
Generating train split: 548304 examples [00:09, 52520.13 examples/s]
Generating train split: 556102 examples [00:09, 56737.07 examples/s]
Generating train split: 563922 examples [00:
+1: 09, 60621.44 examples/s]
Generating train split: 571703 examples [00:09, 63599.65 examples/s]
Generating train split: 579638 examples [00:10, 65868.97 examples/s]
Generating train split: 587480 examples [00:10, 66394.68 examples/s]
Generating train split: 595304 examples [00:10, 68254.66 examples/s]
Generating train split: 602989 examples [00:10, 69394.58 examples/s]
Generating train split: 610780 examples [00:10, 63179.37 examples/s]
Generating train split: 617583 examples [00:10, 51168.73 examples/s]
Generating train split: 625388 examples [00:10, 55681.47 examples/s]
Generating train split: 633184 examples [00:10, 58864.02 examples/s]
Generating train split: 640985 examples [00:11, 62152.72 examples/s]
Generating train split: 648785 examples [00:11, 64277.39 examples/s]
Generating train split: 656749 examples [00:11, 66741.07 examples/s]
Generating train split: 664453 examples [00:11, 68129.37 examples/s]
Generating train split: 672263 examples [00:11, 69100.11 examples/s]
Generating train split: 680061 ex
+1: amples [00:11, 69601.24 examples/s]
Generating train split: 687919 examples [00:11, 71008.69 examples/s]
Generating train split: 695763 examples [00:11, 71968.61 examples/s]
Generating train split: 703538 examples [00:12, 44377.56 examples/s]
Generating train split: 711379 examples [00:12, 49182.06 examples/s]
Generating train split: 719231 examples [00:12, 54252.53 examples/s]
Generating train split: 727015 examples [00:12, 57780.36 examples/s]
Generating train split: 734897 examples [00:12, 61188.71 examples/s]
Generating train split: 742742 examples [00:12, 63899.26 examples/s]
Generating train split: 750481 examples [00:12, 66138.71 examples/s]
Generating train split: 758305 examples [00:12, 67352.77 examples/s]
Generating train split: 765879 examples [00:13, 68167.68 examples/s]
Generating train split: 773830 examples [00:13, 70194.33 examples/s]
Generating train split: 781683 examples [00:13, 70636.63 examples/s]
Generating train split: 788879 examples [00:13, 46262.22 examples/s]
Generating train split
+1: : 795679 examples [00:13, 48649.37 examples/s]
Generating train split: 803406 examples [00:13, 53515.87 examples/s]
Generating train split: 811147 examples [00:13, 57577.18 examples/s]
Generating train split: 818917 examples [00:13, 61191.41 examples/s]
Generating train split: 826743 examples [00:14, 63408.94 examples/s]
Generating train split: 834406 examples [00:14, 64668.32 examples/s]
Generating train split: 842254 examples [00:14, 66933.31 examples/s]
Generating train split: 850052 examples [00:14, 68365.16 examples/s]
Generating train split: 857878 examples [00:14, 70455.32 examples/s]
Generating train split: 865502 examples [00:14, 70709.86 examples/s]
Generating train split: 873144 examples [00:14, 53803.97 examples/s]
Generating train split: 880005 examples [00:15, 47376.41 examples/s]
Generating train split: 887776 examples [00:15, 52582.14 examples/s]
Generating train split: 895507 examples [00:15, 57045.46 examples/s]
Generating train split: 903289 examples [00:15, 60670.78 examples/s]
Generating 
+1: train split: 911125 examples [00:15, 63886.01 examples/s]
Generating train split: 919031 examples [00:15, 66576.69 examples/s]
Generating train split: 926891 examples [00:15, 67221.44 examples/s]
Generating train split: 934726 examples [00:15, 68611.22 examples/s]
Generating train split: 942474 examples [00:15, 69740.25 examples/s]
Generating train split: 950366 examples [00:16, 71460.31 examples/s]
Generating train split: 958225 examples [00:16, 72659.25 examples/s]
Generating train split: 966191 examples [00:16, 49764.06 examples/s]
Generating train split: 973985 examples [00:16, 54015.00 examples/s]
Generating train split: 981624 examples [00:16, 57668.37 examples/s]
Generating train split: 989600 examples [00:16, 61254.75 examples/s]
Generating train split: 997393 examples [00:16, 63930.78 examples/s]
Generating train split: 1005281 examples [00:16, 65881.37 examples/s]
Generating train split: 1012833 examples [00:17, 66702.24 examples/s]
Generating train split: 1020678 examples [00:17, 66810.74 examples/
+1: s]
Generating train split: 1028498 examples [00:17, 68197.21 examples/s]
Generating train split: 1036288 examples [00:17, 70032.83 examples/s]
Generating train split: 1044100 examples [00:17, 71794.98 examples/s]
Generating train split: 1052170 examples [00:18, 31822.82 examples/s]
Generating train split: 1059960 examples [00:18, 29805.55 examples/s]
Generating train split: 1066780 examples [00:18, 34097.75 examples/s]
Generating train split: 1074490 examples [00:18, 40543.54 examples/s]
Generating train split: 1082305 examples [00:18, 46883.99 examples/s]
Generating train split: 1091982 examples [00:18, 49844.45 examples/s]
Generating train split: 1099862 examples [00:18, 54487.30 examples/s]
Generating train split: 1107624 examples [00:19, 58516.62 examples/s]
Generating train split: 1115350 examples [00:19, 61645.73 examples/s]
Generating train split: 1123173 examples [00:19, 65016.49 examples/s]
Generating train split: 1133433 examples [00:19, 50680.96 examples/s]
Generating train split: 1142243 examples 
+1: [00:19, 38864.50 examples/s]
Generating train split: 1149964 examples [00:20, 44358.90 examples/s]
Generating train split: 1157792 examples [00:20, 49800.89 examples/s]
Generating train split: 1165516 examples [00:20, 54287.39 examples/s]
Generating train split: 1173189 examples [00:20, 57988.99 examples/s]
Generating train split: 1180923 examples [00:20, 61256.95 examples/s]
Generating train split: 1188679 examples [00:20, 63888.61 examples/s]
Generating train split: 1196610 examples [00:20, 66174.79 examples/s]
Generating train split: 1204343 examples [00:20, 67675.57 examples/s]
Generating train split: 1212255 examples [00:20, 68336.99 examples/s]
Generating train split: 1220541 examples [00:21, 50532.78 examples/s]
Generating train split: 1230271 examples [00:21, 45068.43 examples/s]
Generating train split: 1238132 examples [00:21, 50277.91 examples/s]
Generating train split: 1245834 examples [00:21, 54452.39 examples/s]
Generating train split: 1253576 examples [00:21, 58238.79 examples/s]
Generating trai
+1: n split: 1261411 examples [00:21, 61327.40 examples/s]
Generating train split: 1269211 examples [00:22, 64210.79 examples/s]
Generating train split: 1277091 examples [00:22, 66183.79 examples/s]
Generating train split: 1284858 examples [00:22, 67589.76 examples/s]
Generating train split: 1292668 examples [00:22, 69671.92 examples/s]
Generating train split: 1300470 examples [00:22, 71475.70 examples/s]
Generating train split: 1308599 examples [00:22, 46959.46 examples/s]
Generating train split: 1318336 examples [00:22, 48384.06 examples/s]
Generating train split: 1326033 examples [00:23, 52828.99 examples/s]
Generating train split: 1333863 examples [00:23, 57088.78 examples/s]
Generating train split: 1341672 examples [00:23, 60635.36 examples/s]
Generating train split: 1349522 examples [00:23, 63399.53 examples/s]
Generating train split: 1357164 examples [00:23, 64575.73 examples/s]
Generating train split: 1365064 examples [00:23, 66833.86 examples/s]
Generating train split: 1372855 examples [00:23, 67960.34 e
+1: xamples/s]
Generating train split: 1380778 examples [00:23, 69800.03 examples/s]
Generating train split: 1388578 examples [00:23, 70972.85 examples/s]
Generating train split: 1393783 examples [00:23, 58119.93 examples/s]
+1: [2025-08-06 23:16:42,514] [INFO] [axolotl.utils.data.wrappers.get_dataset_wrapper:88] [PID:2862582] [RANK:0] Loading dataset: /lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking with base_type: chat_template and prompt_style: None
+1: 
Tokenizing Prompts (num_proc=192):   0%|          | 0/1393783 [00:00<?, ? examples/s]
Tokenizing Prompts (num_proc=192):   0%|          | 1000/1393783 [00:17<6:43:07, 57.58 examples/s]
Tokenizing Prompts (num_proc=192):   0%|          | 2000/1393783 [00:17<2:48:05, 137.99 examples/s]
Tokenizing Prompts (num_proc=192):   0%|          | 3000/1393783 [00:17<1:33:01, 249.17 examples/s]
Tokenizing Prompts (num_proc=192):   0%|          | 4000/1393783 [00:18<1:02:17, 371.90 examples/s]
Tokenizing Prompts (num_proc=192):   0%|          | 5000/1393783 [00:18<41:09, 562.37 examples/s]  
Tokenizing Prompts (num_proc=192):   0%|          | 6000/1393783 [00:18<29:44, 777.61 examples/s]
Tokenizing Prompts (num_proc=192):   1%|          | 7000/1393783 [00:19<21:14, 1087.91 examples/s]
Tokenizing Prompts (num_proc=192):   1%|          | 9000/1393783 [00:19<12:48, 1802.54 examples/s]
Tokenizing Prompts (num_proc=192):   1%|          | 10000/1393783 [00:19<10:28, 2202.45 examples/s]
Tokenizing Prompts (num_proc=192):   1%|  
+1:         | 11000/1393783 [00:19<09:40, 2380.69 examples/s]
Tokenizing Prompts (num_proc=192):   1%|          | 12000/1393783 [00:20<10:11, 2259.42 examples/s]
Tokenizing Prompts (num_proc=192):   1%|          | 13000/1393783 [00:20<08:21, 2750.75 examples/s]
Tokenizing Prompts (num_proc=192):   1%|          | 14000/1393783 [00:20<07:04, 3249.36 examples/s]
Tokenizing Prompts (num_proc=192):   1%|          | 15000/1393783 [00:20<07:13, 3177.87 examples/s]
Tokenizing Prompts (num_proc=192):   1%|          | 17000/1393783 [00:21<05:40, 4047.50 examples/s]
Tokenizing Prompts (num_proc=192):   1%|▏         | 19000/1393783 [00:21<04:56, 4644.42 examples/s]
Tokenizing Prompts (num_proc=192):   1%|▏         | 20000/1393783 [00:21<04:40, 4894.02 examples/s]
Tokenizing Prompts (num_proc=192):   2%|▏         | 21000/1393783 [00:22<04:30, 5076.76 examples/s]
Tokenizing Prompts (num_proc=192):   2%|▏         | 23000/1393783 [00:22<03:26, 6632.43 examples/s]
Tokenizing Prompts (num_proc=192):   2%|▏         | 2400
+1: 0/1393783 [00:22<04:26, 5140.35 examples/s]
Tokenizing Prompts (num_proc=192):   2%|▏         | 25000/1393783 [00:22<04:25, 5161.32 examples/s]
Tokenizing Prompts (num_proc=192):   2%|▏         | 26000/1393783 [00:22<04:25, 5144.56 examples/s]
Tokenizing Prompts (num_proc=192):   2%|▏         | 28000/1393783 [00:23<03:27, 6571.68 examples/s]
Tokenizing Prompts (num_proc=192):   2%|▏         | 29000/1393783 [00:23<03:39, 6220.03 examples/s]
Tokenizing Prompts (num_proc=192):   2%|▏         | 31000/1393783 [00:23<03:55, 5797.17 examples/s]
Tokenizing Prompts (num_proc=192):   2%|▏         | 34000/1393783 [00:23<02:46, 8164.10 examples/s]
Tokenizing Prompts (num_proc=192):   3%|▎         | 35000/1393783 [00:24<03:47, 5970.74 examples/s]
Tokenizing Prompts (num_proc=192):   3%|▎         | 36000/1393783 [00:24<03:55, 5775.72 examples/s]
Tokenizing Prompts (num_proc=192):   3%|▎         | 37000/1393783 [00:24<05:58, 3788.63 examples/s]
Tokenizing Prompts (num_proc=192):   3%|▎         | 38000/13
+1: 93783 [00:25<05:31, 4086.73 examples/s]
Tokenizing Prompts (num_proc=192):   3%|▎         | 39000/1393783 [00:25<05:15, 4290.38 examples/s]
Tokenizing Prompts (num_proc=192):   3%|▎         | 40000/1393783 [00:25<05:48, 3883.39 examples/s]
Tokenizing Prompts (num_proc=192):   3%|▎         | 42000/1393783 [00:25<04:38, 4859.31 examples/s]
Tokenizing Prompts (num_proc=192):   3%|▎         | 43000/1393783 [00:26<06:19, 3558.25 examples/s]
Tokenizing Prompts (num_proc=192):   3%|▎         | 46000/1393783 [00:26<03:47, 5921.78 examples/s]
Tokenizing Prompts (num_proc=192):   3%|▎         | 47000/1393783 [00:27<05:39, 3967.15 examples/s]
Tokenizing Prompts (num_proc=192):   4%|▎         | 51000/1393783 [00:27<03:14, 6907.09 examples/s]
Tokenizing Prompts (num_proc=192):   4%|▍         | 53000/1393783 [00:27<02:41, 8322.75 examples/s]
Tokenizing Prompts (num_proc=192):   4%|▍         | 55000/1393783 [00:27<02:50, 7869.40 examples/s]
Tokenizing Prompts (num_proc=192):   4%|▍         | 57000/139378
+1: 3 [00:28<03:38, 6118.61 examples/s]
Tokenizing Prompts (num_proc=192):   4%|▍         | 59000/1393783 [00:28<03:28, 6406.11 examples/s]
Tokenizing Prompts (num_proc=192):   4%|▍         | 62000/1393783 [00:29<03:15, 6807.25 examples/s]
Tokenizing Prompts (num_proc=192):   5%|▍         | 63000/1393783 [00:29<03:32, 6265.12 examples/s]
Tokenizing Prompts (num_proc=192):   5%|▍         | 64000/1393783 [00:29<04:22, 5070.58 examples/s]
Tokenizing Prompts (num_proc=192):   5%|▍         | 65000/1393783 [00:30<05:18, 4170.33 examples/s]
Tokenizing Prompts (num_proc=192):   5%|▍         | 67000/1393783 [00:30<06:24, 3454.15 examples/s]
Tokenizing Prompts (num_proc=192):   5%|▍         | 69000/1393783 [00:31<06:18, 3498.90 examples/s]
Tokenizing Prompts (num_proc=192):   5%|▌         | 70000/1393783 [00:31<05:52, 3751.05 examples/s]
Tokenizing Prompts (num_proc=192):   5%|▌         | 72000/1393783 [00:31<04:15, 5171.94 examples/s]
Tokenizing Prompts (num_proc=192):   5%|▌         | 73000/1393783 [0
+1: 0:31<04:40, 4712.74 examples/s]
Tokenizing Prompts (num_proc=192):   5%|▌         | 74000/1393783 [00:32<04:51, 4532.01 examples/s]
Tokenizing Prompts (num_proc=192):   6%|▌         | 78000/1393783 [00:32<03:43, 5885.21 examples/s]
Tokenizing Prompts (num_proc=192):   6%|▌         | 79000/1393783 [00:33<04:14, 5166.12 examples/s]
Tokenizing Prompts (num_proc=192):   6%|▌         | 80000/1393783 [00:34<09:31, 2298.76 examples/s]
Tokenizing Prompts (num_proc=192):   6%|▌         | 82000/1393783 [00:34<06:55, 3156.73 examples/s]
Tokenizing Prompts (num_proc=192):   6%|▌         | 83000/1393783 [00:34<05:58, 3655.35 examples/s]
Tokenizing Prompts (num_proc=192):   6%|▌         | 84000/1393783 [00:35<06:04, 3593.96 examples/s]
Tokenizing Prompts (num_proc=192):   6%|▌         | 85000/1393783 [00:35<08:12, 2655.99 examples/s]
Tokenizing Prompts (num_proc=192):   6%|▌         | 86000/1393783 [00:35<07:19, 2975.24 examples/s]
Tokenizing Prompts (num_proc=192):   6%|▌         | 87000/1393783 [00:36
+1: <07:39, 2844.55 examples/s]
Tokenizing Prompts (num_proc=192):   6%|▋         | 89000/1393783 [00:36<06:17, 3457.73 examples/s]
Tokenizing Prompts (num_proc=192):   6%|▋         | 90000/1393783 [00:36<05:39, 3839.98 examples/s]
Tokenizing Prompts (num_proc=192):   7%|▋         | 91000/1393783 [00:37<06:51, 3163.23 examples/s]
Tokenizing Prompts (num_proc=192):   7%|▋         | 92000/1393783 [00:38<08:53, 2441.17 examples/s]
Tokenizing Prompts (num_proc=192):   7%|▋         | 95000/1393783 [00:38<07:20, 2951.42 examples/s]
Tokenizing Prompts (num_proc=192):   7%|▋         | 97000/1393783 [00:39<08:16, 2613.72 examples/s]
Tokenizing Prompts (num_proc=192):   7%|▋         | 98000/1393783 [00:40<08:42, 2478.64 examples/s]
Tokenizing Prompts (num_proc=192):   7%|▋         | 99000/1393783 [00:40<09:09, 2356.53 examples/s]
Tokenizing Prompts (num_proc=192):   7%|▋         | 100000/1393783 [00:41<10:16, 2098.62 examples/s]
Tokenizing Prompts (num_proc=192):   7%|▋         | 103000/1393783 [00:41<0
+1: 6:42, 3209.35 examples/s]
Tokenizing Prompts (num_proc=192):   8%|▊         | 105000/1393783 [00:42<04:58, 4318.50 examples/s]
Tokenizing Prompts (num_proc=192):   8%|▊         | 107000/1393783 [00:42<04:07, 5206.68 examples/s]
Tokenizing Prompts (num_proc=192):   8%|▊         | 108000/1393783 [00:42<04:03, 5285.83 examples/s]
Tokenizing Prompts (num_proc=192):   8%|▊         | 109000/1393783 [00:42<03:47, 5653.18 examples/s]
Tokenizing Prompts (num_proc=192):   8%|▊         | 111000/1393783 [00:42<03:15, 6566.51 examples/s]
Tokenizing Prompts (num_proc=192):   8%|▊         | 115000/1393783 [00:43<02:01, 10531.53 examples/s]
Tokenizing Prompts (num_proc=192):   8%|▊         | 117000/1393783 [00:43<02:11, 9675.49 examples/s] 
Tokenizing Prompts (num_proc=192):   9%|▊         | 119000/1393783 [00:43<03:33, 5957.74 examples/s]
Tokenizing Prompts (num_proc=192):   9%|▊         | 120000/1393783 [00:44<03:49, 5546.57 examples/s]
Tokenizing Prompts (num_proc=192):   9%|▉         | 123000/1393783 
+1: [00:44<02:51, 7425.82 examples/s]
Tokenizing Prompts (num_proc=192):   9%|▉         | 125000/1393783 [00:44<02:31, 8365.53 examples/s]
Tokenizing Prompts (num_proc=192):   9%|▉         | 127000/1393783 [00:44<02:21, 8982.23 examples/s]
Tokenizing Prompts (num_proc=192):   9%|▉         | 129000/1393783 [00:45<02:37, 8042.35 examples/s]
Tokenizing Prompts (num_proc=192):   9%|▉         | 130000/1393783 [00:45<04:53, 4299.81 examples/s]
Tokenizing Prompts (num_proc=192):   9%|▉         | 132000/1393783 [00:45<03:46, 5573.92 examples/s]
Tokenizing Prompts (num_proc=192):  10%|▉         | 133000/1393783 [00:46<05:03, 4148.05 examples/s]
Tokenizing Prompts (num_proc=192):  10%|▉         | 135000/1393783 [00:46<03:45, 5591.30 examples/s]
Tokenizing Prompts (num_proc=192):  10%|▉         | 136000/1393783 [00:46<03:53, 5389.37 examples/s]
Tokenizing Prompts (num_proc=192):  10%|▉         | 137000/1393783 [00:46<03:31, 5953.48 examples/s]
Tokenizing Prompts (num_proc=192):  10%|▉         | 138000/13
+1: 93783 [00:47<03:45, 5575.32 examples/s]
Tokenizing Prompts (num_proc=192):  10%|█         | 140000/1393783 [00:47<03:02, 6874.90 examples/s]
Tokenizing Prompts (num_proc=192):  10%|█         | 142000/1393783 [00:47<02:28, 8446.82 examples/s]
Tokenizing Prompts (num_proc=192):  10%|█         | 145000/1393783 [00:47<02:07, 9799.26 examples/s]
Tokenizing Prompts (num_proc=192):  11%|█         | 147000/1393783 [00:48<02:35, 8014.70 examples/s]
Tokenizing Prompts (num_proc=192):  11%|█         | 148000/1393783 [00:48<03:26, 6023.98 examples/s]
Tokenizing Prompts (num_proc=192):  11%|█         | 150000/1393783 [00:48<03:07, 6648.22 examples/s]
Tokenizing Prompts (num_proc=192):  11%|█         | 151000/1393783 [00:49<04:14, 4890.46 examples/s]
Tokenizing Prompts (num_proc=192):  11%|█         | 152000/1393783 [00:49<05:07, 4033.17 examples/s]
Tokenizing Prompts (num_proc=192):  11%|█         | 153000/1393783 [00:49<05:18, 3889.66 examples/s]
Tokenizing Prompts (num_proc=192):  11%|█         | 154
+1: 000/1393783 [00:49<04:31, 4559.39 examples/s]
Tokenizing Prompts (num_proc=192):  11%|█         | 156000/1393783 [00:50<03:55, 5266.37 examples/s]
Tokenizing Prompts (num_proc=192):  11%|█▏        | 159000/1393783 [00:50<03:01, 6797.16 examples/s]
Tokenizing Prompts (num_proc=192):  11%|█▏        | 160000/1393783 [00:50<03:20, 6167.35 examples/s]
Tokenizing Prompts (num_proc=192):  12%|█▏        | 162000/1393783 [00:50<02:39, 7739.36 examples/s]
Tokenizing Prompts (num_proc=192):  12%|█▏        | 164000/1393783 [00:51<02:37, 7789.40 examples/s]
Tokenizing Prompts (num_proc=192):  12%|█▏        | 165000/1393783 [00:51<02:41, 7597.74 examples/s]
Tokenizing Prompts (num_proc=192):  12%|█▏        | 166000/1393783 [00:51<02:46, 7366.06 examples/s]
Tokenizing Prompts (num_proc=192):  12%|█▏        | 170000/1393783 [00:51<01:50, 11033.84 examples/s]
Tokenizing Prompts (num_proc=192):  12%|█▏        | 172000/1393783 [00:52<02:50, 7148.81 examples/s] 
Tokenizing Prompts (num_proc=192)
+1: :  13%|█▎        | 175000/1393783 [00:52<02:07, 9527.36 examples/s]
Tokenizing Prompts (num_proc=192):  13%|█▎        | 177000/1393783 [00:52<01:57, 10363.06 examples/s]
Tokenizing Prompts (num_proc=192):  13%|█▎        | 180000/1393783 [00:52<02:03, 9811.39 examples/s] 
Tokenizing Prompts (num_proc=192):  13%|█▎        | 182000/1393783 [00:52<02:00, 10094.31 examples/s]
Tokenizing Prompts (num_proc=192):  13%|█▎        | 184000/1393783 [00:53<02:29, 8106.91 examples/s] 
Tokenizing Prompts (num_proc=192):  13%|█▎        | 185000/1393783 [00:53<02:37, 7685.32 examples/s]
Tokenizing Prompts (num_proc=192):  13%|█▎        | 186000/1393783 [00:53<03:31, 5698.46 examples/s]
Tokenizing Prompts (num_proc=192):  13%|█▎        | 187000/1393783 [00:54<03:27, 5808.77 examples/s]
Tokenizing Prompts (num_proc=192):  14%|█▎        | 189000/1393783 [00:54<03:14, 6205.88 examples/s]
Tokenizing Prompts (num_proc=192):  14%|█▎        | 190000/1393783 [00:54<03:30, 5728.83 examples/s]
Tok
+1: enizing Prompts (num_proc=192):  14%|█▍        | 193000/1393783 [00:54<02:16, 8797.61 examples/s]
Tokenizing Prompts (num_proc=192):  14%|█▍        | 195000/1393783 [00:54<02:27, 8116.49 examples/s]
Tokenizing Prompts (num_proc=192):  14%|█▍        | 196000/1393783 [00:55<02:39, 7523.00 examples/s]
Tokenizing Prompts (num_proc=192):  14%|█▍        | 197000/1393783 [00:55<03:12, 6217.05 examples/s]
Tokenizing Prompts (num_proc=192):  14%|█▍        | 198000/1393783 [00:55<03:05, 6448.27 examples/s]
Tokenizing Prompts (num_proc=192):  14%|█▍        | 199000/1393783 [00:55<03:21, 5933.10 examples/s]
Tokenizing Prompts (num_proc=192):  14%|█▍        | 200000/1393783 [00:55<03:33, 5603.11 examples/s]
Tokenizing Prompts (num_proc=192):  14%|█▍        | 201000/1393783 [00:56<03:16, 6071.43 examples/s]
Tokenizing Prompts (num_proc=192):  15%|█▍        | 203000/1393783 [00:56<02:34, 7723.85 examples/s]
Tokenizing Prompts (num_proc=192):  15%|█▍        | 204000/1393783 [00:56<02:4
+1: 0, 7410.91 examples/s]
Tokenizing Prompts (num_proc=192):  15%|█▍        | 206000/1393783 [00:56<02:11, 9041.08 examples/s]
Tokenizing Prompts (num_proc=192):  15%|█▍        | 208000/1393783 [00:56<02:14, 8792.15 examples/s]
Tokenizing Prompts (num_proc=192):  15%|█▍        | 209000/1393783 [00:56<02:30, 7864.94 examples/s]
Tokenizing Prompts (num_proc=192):  15%|█▌        | 212000/1393783 [00:57<01:40, 11812.62 examples/s]
Tokenizing Prompts (num_proc=192):  15%|█▌        | 214000/1393783 [00:57<02:17, 8561.31 examples/s] 
Tokenizing Prompts (num_proc=192):  16%|█▌        | 217000/1393783 [00:57<01:49, 10763.68 examples/s]
Tokenizing Prompts (num_proc=192):  16%|█▌        | 219000/1393783 [00:58<02:30, 7789.92 examples/s] 
Tokenizing Prompts (num_proc=192):  16%|█▌        | 222000/1393783 [00:58<02:02, 9535.98 examples/s]
Tokenizing Prompts (num_proc=192):  16%|█▌        | 224000/1393783 [00:58<01:52, 10409.27 examples/s]
Tokenizing Prompts (num_proc=192):  16%|█▋     
+1:    | 227000/1393783 [00:59<02:56, 6595.46 examples/s] 
Tokenizing Prompts (num_proc=192):  16%|█▋        | 229000/1393783 [00:59<02:49, 6885.43 examples/s]
Tokenizing Prompts (num_proc=192):  17%|█▋        | 230000/1393783 [00:59<02:53, 6695.22 examples/s]
Tokenizing Prompts (num_proc=192):  17%|█▋        | 231000/1393783 [00:59<02:51, 6762.60 examples/s]
Tokenizing Prompts (num_proc=192):  17%|█▋        | 232000/1393783 [00:59<02:55, 6632.36 examples/s]
Tokenizing Prompts (num_proc=192):  17%|█▋        | 235000/1393783 [00:59<01:55, 10072.77 examples/s]
Tokenizing Prompts (num_proc=192):  17%|█▋        | 237000/1393783 [01:00<02:38, 7299.91 examples/s] 
Tokenizing Prompts (num_proc=192):  17%|█▋        | 239000/1393783 [01:00<02:41, 7170.32 examples/s]
Tokenizing Prompts (num_proc=192):  17%|█▋        | 240000/1393783 [01:00<02:37, 7336.74 examples/s]
Tokenizing Prompts (num_proc=192):  17%|█▋        | 241000/1393783 [01:00<02:29, 7709.15 examples/s]
Tokenizing Prompts (nu
+1: m_proc=192):  17%|█▋        | 242000/1393783 [01:01<02:39, 7233.41 examples/s]
Tokenizing Prompts (num_proc=192):  18%|█▊        | 244000/1393783 [01:01<02:36, 7333.41 examples/s]
Tokenizing Prompts (num_proc=192):  18%|█▊        | 246000/1393783 [01:02<04:47, 3995.46 examples/s]
Tokenizing Prompts (num_proc=192):  18%|█▊        | 248000/1393783 [01:02<04:13, 4512.02 examples/s]
Tokenizing Prompts (num_proc=192):  18%|█▊        | 250000/1393783 [01:02<03:28, 5480.36 examples/s]
Tokenizing Prompts (num_proc=192):  18%|█▊        | 252000/1393783 [01:03<03:22, 5642.20 examples/s]
Tokenizing Prompts (num_proc=192):  18%|█▊        | 253000/1393783 [01:03<03:44, 5084.64 examples/s]
Tokenizing Prompts (num_proc=192):  18%|█▊        | 254000/1393783 [01:03<03:35, 5293.72 examples/s]
Tokenizing Prompts (num_proc=192):  18%|█▊        | 255000/1393783 [01:04<05:47, 3279.31 examples/s]
Tokenizing Prompts (num_proc=192):  18%|█▊        | 256000/1393783 [01:04<05:28, 3462.27 examples
+1: /s]
Tokenizing Prompts (num_proc=192):  19%|█▊        | 258000/1393783 [01:04<04:22, 4325.53 examples/s]
Tokenizing Prompts (num_proc=192):  19%|█▊        | 260000/1393783 [01:05<06:29, 2910.89 examples/s]
Tokenizing Prompts (num_proc=192):  19%|█▉        | 262000/1393783 [01:06<05:11, 3628.39 examples/s]
Tokenizing Prompts (num_proc=192):  19%|█▉        | 263000/1393783 [01:07<07:44, 2435.87 examples/s]
Tokenizing Prompts (num_proc=192):  19%|█▉        | 265000/1393783 [01:07<05:48, 3240.66 examples/s]
Tokenizing Prompts (num_proc=192):  19%|█▉        | 267000/1393783 [01:07<04:35, 4090.27 examples/s]
Tokenizing Prompts (num_proc=192):  19%|█▉        | 269000/1393783 [01:07<03:54, 4800.25 examples/s]
Tokenizing Prompts (num_proc=192):  19%|█▉        | 270000/1393783 [01:08<05:28, 3418.00 examples/s]
Tokenizing Prompts (num_proc=192):  19%|█▉        | 271000/1393783 [01:08<05:38, 3320.59 examples/s]
Tokenizing Prompts (num_proc=192):  20%|█▉        | 272000/1393783 [01:
+1: 09<05:34, 3354.52 examples/s]
Tokenizing Prompts (num_proc=192):  20%|█▉        | 273000/1393783 [01:10<08:53, 2100.99 examples/s]
Tokenizing Prompts (num_proc=192):  20%|█▉        | 274000/1393783 [01:10<08:18, 2246.09 examples/s]
Tokenizing Prompts (num_proc=192):  20%|█▉        | 277000/1393783 [01:10<04:22, 4246.83 examples/s]
Tokenizing Prompts (num_proc=192):  20%|█▉        | 278000/1393783 [01:11<04:53, 3800.48 examples/s]
Tokenizing Prompts (num_proc=192):  20%|██        | 279000/1393783 [01:11<04:28, 4153.10 examples/s]
Tokenizing Prompts (num_proc=192):  20%|██        | 281000/1393783 [01:11<04:13, 4393.36 examples/s]
Tokenizing Prompts (num_proc=192):  20%|██        | 282000/1393783 [01:11<03:57, 4672.39 examples/s]
Tokenizing Prompts (num_proc=192):  20%|██        | 283000/1393783 [01:12<03:56, 4704.62 examples/s]
Tokenizing Prompts (num_proc=192):  20%|██        | 285000/1393783 [01:12<03:33, 5192.09 examples/s]
Tokenizing Prompts (num_proc=192):  21%|██   
+1:      | 286000/1393783 [01:13<06:25, 2871.27 examples/s]
Tokenizing Prompts (num_proc=192):  21%|██        | 287000/1393783 [01:13<05:30, 3352.34 examples/s]
Tokenizing Prompts (num_proc=192):  21%|██        | 288000/1393783 [01:13<05:02, 3649.83 examples/s]
Tokenizing Prompts (num_proc=192):  21%|██        | 293000/1393783 [01:13<02:15, 8145.64 examples/s]
Tokenizing Prompts (num_proc=192):  21%|██        | 295000/1393783 [01:14<02:19, 7890.60 examples/s]
Tokenizing Prompts (num_proc=192):  21%|██▏       | 297000/1393783 [01:14<03:01, 6044.69 examples/s]
Tokenizing Prompts (num_proc=192):  21%|██▏       | 298000/1393783 [01:14<02:54, 6279.44 examples/s]
Tokenizing Prompts (num_proc=192):  22%|██▏       | 301000/1393783 [01:14<02:18, 7917.56 examples/s]
Tokenizing Prompts (num_proc=192):  22%|██▏       | 302000/1393783 [01:15<02:30, 7244.23 examples/s]
Tokenizing Prompts (num_proc=192):  22%|██▏       | 303000/1393783 [01:15<02:26, 7465.41 examples/s]
Tokenizing Pr
+1: ompts (num_proc=192):  22%|██▏       | 305000/1393783 [01:15<02:47, 6518.44 examples/s]
Tokenizing Prompts (num_proc=192):  22%|██▏       | 307000/1393783 [01:15<02:16, 7950.00 examples/s]
Tokenizing Prompts (num_proc=192):  22%|██▏       | 308000/1393783 [01:15<02:15, 8001.16 examples/s]
Tokenizing Prompts (num_proc=192):  22%|██▏       | 309000/1393783 [01:16<04:13, 4278.38 examples/s]
Tokenizing Prompts (num_proc=192):  22%|██▏       | 310000/1393783 [01:17<05:18, 3398.99 examples/s]
Tokenizing Prompts (num_proc=192):  22%|██▏       | 311000/1393783 [01:17<06:36, 2731.22 examples/s]
Tokenizing Prompts (num_proc=192):  22%|██▏       | 313000/1393783 [01:17<04:20, 4142.61 examples/s]
Tokenizing Prompts (num_proc=192):  23%|██▎       | 314000/1393783 [01:18<06:09, 2926.13 examples/s]
Tokenizing Prompts (num_proc=192):  23%|██▎       | 315000/1393783 [01:18<06:37, 2715.79 examples/s]
Tokenizing Prompts (num_proc=192):  23%|██▎       | 316000/1393783 [
+1: 01:19<05:26, 3304.83 examples/s]
Tokenizing Prompts (num_proc=192):  23%|██▎       | 317000/1393783 [01:19<05:06, 3514.42 examples/s]
Tokenizing Prompts (num_proc=192):  23%|██▎       | 319000/1393783 [01:19<03:24, 5256.42 examples/s]
Tokenizing Prompts (num_proc=192):  23%|██▎       | 320000/1393783 [01:19<04:01, 4450.22 examples/s]
Tokenizing Prompts (num_proc=192):  23%|██▎       | 322000/1393783 [01:19<03:02, 5862.58 examples/s]
Tokenizing Prompts (num_proc=192):  23%|██▎       | 323000/1393783 [01:20<05:37, 3170.51 examples/s]
Tokenizing Prompts (num_proc=192):  23%|██▎       | 324000/1393783 [01:21<05:57, 2992.29 examples/s]
Tokenizing Prompts (num_proc=192):  23%|██▎       | 325000/1393783 [01:21<05:29, 3244.42 examples/s]
Tokenizing Prompts (num_proc=192):  23%|██▎       | 327000/1393783 [01:21<04:27, 3988.07 examples/s]
Tokenizing Prompts (num_proc=192):  24%|██▎       | 328000/1393783 [01:22<04:54, 3614.27 examples/s]
Tokenizing Prompts (num_proc
+1: =192):  24%|██▎       | 331000/1393783 [01:22<02:55, 6071.38 examples/s]
Tokenizing Prompts (num_proc=192):  24%|██▍       | 333000/1393783 [01:22<02:22, 7434.15 examples/s]
Tokenizing Prompts (num_proc=192):  24%|██▍       | 335000/1393783 [01:22<02:48, 6275.68 examples/s]
Tokenizing Prompts (num_proc=192):  24%|██▍       | 337000/1393783 [01:22<02:16, 7714.60 examples/s]
Tokenizing Prompts (num_proc=192):  24%|██▍       | 339000/1393783 [01:23<02:45, 6356.83 examples/s]
Tokenizing Prompts (num_proc=192):  24%|██▍       | 340000/1393783 [01:23<02:46, 6344.33 examples/s]
Tokenizing Prompts (num_proc=192):  24%|██▍       | 341000/1393783 [01:23<02:42, 6459.09 examples/s]
Tokenizing Prompts (num_proc=192):  25%|██▍       | 343000/1393783 [01:23<02:26, 7156.55 examples/s]
Tokenizing Prompts (num_proc=192):  25%|██▍       | 345000/1393783 [01:24<02:42, 6469.00 examples/s]
Tokenizing Prompts (num_proc=192):  25%|██▌       | 349000/1393783 [01:24<01:42, 10
+1: 206.85 examples/s]
Tokenizing Prompts (num_proc=192):  25%|██▌       | 351000/1393783 [01:24<02:00, 8640.17 examples/s] 
Tokenizing Prompts (num_proc=192):  25%|██▌       | 353000/1393783 [01:25<02:27, 7041.80 examples/s]
Tokenizing Prompts (num_proc=192):  26%|██▌       | 356000/1393783 [01:25<01:55, 9009.05 examples/s]
Tokenizing Prompts (num_proc=192):  26%|██▌       | 359000/1393783 [01:25<01:30, 11431.55 examples/s]
Tokenizing Prompts (num_proc=192):  26%|██▌       | 361000/1393783 [01:25<01:42, 10063.55 examples/s]
Tokenizing Prompts (num_proc=192):  26%|██▌       | 363000/1393783 [01:26<02:23, 7159.04 examples/s] 
Tokenizing Prompts (num_proc=192):  26%|██▌       | 365000/1393783 [01:26<02:29, 6904.06 examples/s]
Tokenizing Prompts (num_proc=192):  26%|██▋       | 367000/1393783 [01:26<02:03, 8300.32 examples/s]
Tokenizing Prompts (num_proc=192):  26%|██▋       | 369000/1393783 [01:26<01:58, 8666.96 examples/s]
Tokenizing Prompts (num_proc=192):  27
+1: %|██▋       | 371000/1393783 [01:27<02:19, 7326.52 examples/s]
Tokenizing Prompts (num_proc=192):  27%|██▋       | 372000/1393783 [01:27<03:47, 4485.91 examples/s]
Tokenizing Prompts (num_proc=192):  27%|██▋       | 374000/1393783 [01:28<03:19, 5099.83 examples/s]
Tokenizing Prompts (num_proc=192):  27%|██▋       | 375000/1393783 [01:28<03:30, 4847.43 examples/s]
Tokenizing Prompts (num_proc=192):  27%|██▋       | 377000/1393783 [01:28<02:46, 6114.24 examples/s]
Tokenizing Prompts (num_proc=192):  27%|���█▋       | 380000/1393783 [01:28<01:57, 8649.96 examples/s]
Tokenizing Prompts (num_proc=192):  27%|██▋       | 383000/1393783 [01:28<01:33, 10800.35 examples/s]
Tokenizing Prompts (num_proc=192):  28%|██▊       | 385000/1393783 [01:29<01:36, 10404.22 examples/s]
Tokenizing Prompts (num_proc=192):  28%|██▊       | 387000/1393783 [01:29<01:44, 9653.12 examples/s] 
Tokenizing Prompts (num_proc=192):  28%|██▊       | 389000/1393783 [01:29<01:31, 11001.79 
+1: examples/s]
Tokenizing Prompts (num_proc=192):  28%|██▊       | 391000/1393783 [01:29<01:34, 10595.61 examples/s]
Tokenizing Prompts (num_proc=192):  28%|██▊       | 393000/1393783 [01:29<01:48, 9208.47 examples/s] 
Tokenizing Prompts (num_proc=192):  28%|██▊       | 395000/1393783 [01:30<01:39, 10016.81 examples/s]
Tokenizing Prompts (num_proc=192):  28%|██▊       | 397000/1393783 [01:30<01:38, 10081.86 examples/s]
Tokenizing Prompts (num_proc=192):  29%|██▊       | 399000/1393783 [01:30<01:25, 11640.44 examples/s]
Tokenizing Prompts (num_proc=192):  29%|██▉       | 401000/1393783 [01:30<01:35, 10398.26 examples/s]
Tokenizing Prompts (num_proc=192):  29%|██▉       | 403000/1393783 [01:30<01:44, 9436.13 examples/s] 
Tokenizing Prompts (num_proc=192):  29%|██▉       | 406000/1393783 [01:31<01:50, 8900.09 examples/s]
Tokenizing Prompts (num_proc=192):  29%|██▉       | 407000/1393783 [01:31<02:31, 6526.89 examples/s]
Tokenizing Prompts (num_proc=192):  29%|�
+1: �█▉       | 408000/1393783 [01:31<02:27, 6663.48 examples/s]
Tokenizing Prompts (num_proc=192):  29%|██▉       | 410000/1393783 [01:32<02:57, 5556.09 examples/s]
Tokenizing Prompts (num_proc=192):  29%|██▉       | 411000/1393783 [01:32<03:02, 5388.73 examples/s]
Tokenizing Prompts (num_proc=192):  30%|██▉       | 413000/1393783 [01:32<02:25, 6734.45 examples/s]
Tokenizing Prompts (num_proc=192):  30%|██▉       | 415000/1393783 [01:32<01:55, 8485.63 examples/s]
Tokenizing Prompts (num_proc=192):  30%|██▉       | 417000/1393783 [01:33<01:54, 8559.65 examples/s]
Tokenizing Prompts (num_proc=192):  30%|███       | 419000/1393783 [01:33<02:34, 6294.79 examples/s]
Tokenizing Prompts (num_proc=192):  30%|███       | 420000/1393783 [01:33<03:06, 5215.99 examples/s]
Tokenizing Prompts (num_proc=192):  30%|███       | 421000/1393783 [01:34<03:16, 4938.01 examples/s]
Tokenizing Prompts (num_proc=192):  30%|███       | 422000/1393783 [01:34<03:18, 4898.05 examples
+1: /s]
Tokenizing Prompts (num_proc=192):  30%|███       | 423000/1393783 [01:34<03:15, 4974.83 examples/s]
Tokenizing Prompts (num_proc=192):  30%|███       | 424000/1393783 [01:34<02:50, 5696.87 examples/s]
Tokenizing Prompts (num_proc=192):  31%|███       | 426000/1393783 [01:34<02:06, 7660.32 examples/s]
Tokenizing Prompts (num_proc=192):  31%|███       | 427000/1393783 [01:34<02:33, 6309.96 examples/s]
Tokenizing Prompts (num_proc=192):  31%|███       | 428000/1393783 [01:35<05:46, 2789.89 examples/s]
Tokenizing Prompts (num_proc=192):  31%|███       | 430000/1393783 [01:36<04:42, 3416.56 examples/s]
Tokenizing Prompts (num_proc=192):  31%|███       | 431000/1393783 [01:36<04:30, 3560.13 examples/s]
Tokenizing Prompts (num_proc=192):  31%|███       | 433000/1393783 [01:37<05:00, 3193.47 examples/s]
Tokenizing Prompts (num_proc=192):  31%|███       | 434000/1393783 [01:37<04:20, 3685.36 examples/s]
Tokenizing Prompts (num_proc=192):  31%|███       |
+1:  435000/1393783 [01:37<04:04, 3916.76 examples/s]
Tokenizing Prompts (num_proc=192):  31%|███▏      | 436000/1393783 [01:38<04:49, 3313.00 examples/s]
Tokenizing Prompts (num_proc=192):  31%|███▏      | 437000/1393783 [01:38<05:36, 2840.17 examples/s]
Tokenizing Prompts (num_proc=192):  31%|███▏      | 438000/1393783 [01:38<04:46, 3334.53 examples/s]
Tokenizing Prompts (num_proc=192):  31%|███▏      | 439000/1393783 [01:39<05:09, 3089.66 examples/s]
Tokenizing Prompts (num_proc=192):  32%|███▏      | 440000/1393783 [01:39<06:53, 2307.78 examples/s]
Tokenizing Prompts (num_proc=192):  32%|███▏      | 441000/1393783 [01:39<05:35, 2835.68 examples/s]
Tokenizing Prompts (num_proc=192):  32%|███▏      | 444000/1393783 [01:40<03:27, 4585.28 examples/s]
Tokenizing Prompts (num_proc=192):  32%|███▏      | 445000/1393783 [01:40<03:17, 4803.58 examples/s]
Tokenizing Prompts (num_proc=192):  32%|███▏      | 447000/1393783 [01:40<03:08, 5017.03 examp
+1: les/s]
Tokenizing Prompts (num_proc=192):  32%|███▏      | 448000/1393783 [01:40<02:57, 5333.80 examples/s]
Tokenizing Prompts (num_proc=192):  32%|███▏      | 452000/1393783 [01:41<01:49, 8563.33 examples/s]
Tokenizing Prompts (num_proc=192):  33%|███▎      | 453000/1393783 [01:41<02:41, 5832.35 examples/s]
Tokenizing Prompts (num_proc=192):  33%|███▎      | 454000/1393783 [01:41<02:47, 5601.77 examples/s]
Tokenizing Prompts (num_proc=192):  33%|███▎      | 455000/1393783 [01:42<02:53, 5399.92 examples/s]
Tokenizing Prompts (num_proc=192):  33%|███▎      | 456000/1393783 [01:42<04:56, 3162.59 examples/s]
Tokenizing Prompts (num_proc=192):  33%|███▎      | 457000/1393783 [01:43<04:47, 3253.34 examples/s]
Tokenizing Prompts (num_proc=192):  33%|███▎      | 458000/1393783 [01:43<04:12, 3703.25 examples/s]
Tokenizing Prompts (num_proc=192):  33%|███▎      | 459000/1393783 [01:43<04:03, 3846.20 examples/s]
Tokenizing Prompts (num_proc=192):  
+1: 33%|███▎      | 460000/1393783 [01:43<03:53, 4002.65 examples/s]
Tokenizing Prompts (num_proc=192):  33%|███▎      | 461000/1393783 [01:43<03:35, 4326.87 examples/s]
Tokenizing Prompts (num_proc=192):  33%|███▎      | 463000/1393783 [01:44<04:03, 3827.39 examples/s]
Tokenizing Prompts (num_proc=192):  33%|███▎      | 464000/1393783 [01:44<03:33, 4351.73 examples/s]
Tokenizing Prompts (num_proc=192):  34%|███▎      | 467000/1393783 [01:44<02:35, 5957.02 examples/s]
Tokenizing Prompts (num_proc=192):  34%|███▎      | 468000/1393783 [01:45<03:39, 4209.86 examples/s]
Tokenizing Prompts (num_proc=192):  34%|███▎      | 469000/1393783 [01:45<03:20, 4616.94 examples/s]
Tokenizing Prompts (num_proc=192):  34%|███▍      | 471000/1393783 [01:46<03:22, 4547.57 examples/s]
Tokenizing Prompts (num_proc=192):  34%|███▍      | 472000/1393783 [01:46<03:34, 4293.57 examples/s]
Tokenizing Prompts (num_proc=192):  34%|███▍      | 473000/1393783 [01:
+1: 46<03:12, 4783.46 examples/s]
Tokenizing Prompts (num_proc=192):  34%|███▍      | 474000/1393783 [01:46<02:51, 5361.88 examples/s]
Tokenizing Prompts (num_proc=192):  34%|███▍      | 476000/1393783 [01:46<02:36, 5859.70 examples/s]
Tokenizing Prompts (num_proc=192):  34%|███▍      | 479000/1393783 [01:47<02:01, 7512.26 examples/s]
Tokenizing Prompts (num_proc=192):  34%|███▍      | 480000/1393783 [01:47<02:24, 6330.99 examples/s]
Tokenizing Prompts (num_proc=192):  35%|███▍      | 482000/1393783 [01:47<02:52, 5292.22 examples/s]
Tokenizing Prompts (num_proc=192):  35%|███▍      | 483000/1393783 [01:48<02:44, 5552.29 examples/s]
Tokenizing Prompts (num_proc=192):  35%|███▍      | 484000/1393783 [01:48<02:47, 5439.72 examples/s]
Tokenizing Prompts (num_proc=192):  35%|███▍      | 486000/1393783 [01:48<02:28, 6107.78 examples/s]
Tokenizing Prompts (num_proc=192):  35%|███▍      | 487000/1393783 [01:48<02:34, 5851.44 examples/s]
Tokenizing Pr
+1: ompts (num_proc=192):  35%|███▌      | 488000/1393783 [01:48<02:33, 5885.74 examples/s]
Tokenizing Prompts (num_proc=192):  35%|███▌      | 491000/1393783 [01:49<02:58, 5063.01 examples/s]
Tokenizing Prompts (num_proc=192):  35%|███▌      | 493000/1393783 [01:50<03:50, 3901.37 examples/s]
Tokenizing Prompts (num_proc=192):  36%|███▌      | 495000/1393783 [01:50<03:23, 4421.89 examples/s]
Tokenizing Prompts (num_proc=192):  36%|███▌      | 497000/1393783 [01:51<03:46, 3953.97 examples/s]
Tokenizing Prompts (num_proc=192):  36%|███▌      | 498000/1393783 [01:51<03:29, 4266.05 examples/s]
Tokenizing Prompts (num_proc=192):  36%|███▌      | 499000/1393783 [01:51<03:15, 4573.25 examples/s]
Tokenizing Prompts (num_proc=192):  36%|███▌      | 501000/1393783 [01:51<02:43, 5471.65 examples/s]
Tokenizing Prompts (num_proc=192):  36%|███▌      | 502000/1393783 [01:52<03:01, 4908.31 examples/s]
Tokenizing Prompts (num_proc=192):  36%|███▌    
+1:   | 504000/1393783 [01:52<03:21, 4407.59 examples/s]
Tokenizing Prompts (num_proc=192):  36%|███▌      | 505000/1393783 [01:52<03:39, 4043.01 examples/s]
Tokenizing Prompts (num_proc=192):  36%|███▋      | 507000/1393783 [01:53<02:37, 5627.02 examples/s]
Tokenizing Prompts (num_proc=192):  36%|███▋      | 508000/1393783 [01:53<03:50, 3835.44 examples/s]
Tokenizing Prompts (num_proc=192):  37%|███▋      | 510000/1393783 [01:53<02:50, 5193.14 examples/s]
Tokenizing Prompts (num_proc=192):  37%|███▋      | 512000/1393783 [01:53<02:06, 6993.23 examples/s]
Tokenizing Prompts (num_proc=192):  37%|███▋      | 515000/1393783 [01:54<01:25, 10219.02 examples/s]
Tokenizing Prompts (num_proc=192):  37%|███▋      | 517000/1393783 [01:54<02:06, 6916.77 examples/s] 
Tokenizing Prompts (num_proc=192):  37%|███▋      | 519000/1393783 [01:54<02:01, 7181.57 examples/s]
Tokenizing Prompts (num_proc=192):  37%|███▋      | 521000/1393783 [01:55<02:09, 6754.18 
+1: examples/s]
Tokenizing Prompts (num_proc=192):  37%|███▋      | 522000/1393783 [01:55<02:10, 6698.05 examples/s]
Tokenizing Prompts (num_proc=192):  38%|███▊      | 525000/1393783 [01:55<02:36, 5537.73 examples/s]
Tokenizing Prompts (num_proc=192):  38%|███▊      | 526000/1393783 [01:56<03:06, 4655.22 examples/s]
Tokenizing Prompts (num_proc=192):  38%|███▊      | 527000/1393783 [01:56<03:19, 4342.58 examples/s]
Tokenizing Prompts (num_proc=192):  38%|███▊      | 530000/1393783 [01:56<02:06, 6842.74 examples/s]
Tokenizing Prompts (num_proc=192):  38%|███▊      | 532000/1393783 [01:57<02:09, 6629.28 examples/s]
Tokenizing Prompts (num_proc=192):  38%|███▊      | 534000/1393783 [01:57<01:53, 7563.87 examples/s]
Tokenizing Prompts (num_proc=192):  38%|███▊      | 535000/1393783 [01:57<01:48, 7891.82 examples/s]
Tokenizing Prompts (num_proc=192):  39%|███▊      | 537000/1393783 [01:57<01:37, 8810.48 examples/s]
Tokenizing Prompts (num_proc=19
+1: 2):  39%|███▊      | 539000/1393783 [01:58<02:22, 6001.49 examples/s]
Tokenizing Prompts (num_proc=192):  39%|███▊      | 540000/1393783 [01:58<02:11, 6474.19 examples/s]
Tokenizing Prompts (num_proc=192):  39%|███▉      | 541000/1393783 [01:58<03:01, 4687.65 examples/s]
Tokenizing Prompts (num_proc=192):  39%|███▉      | 542000/1393783 [01:58<03:00, 4726.83 examples/s]
Tokenizing Prompts (num_proc=192):  39%|███▉      | 543000/1393783 [01:59<02:54, 4880.58 examples/s]
Tokenizing Prompts (num_proc=192):  39%|███▉      | 544000/1393783 [01:59<03:32, 4004.31 examples/s]
Tokenizing Prompts (num_proc=192):  39%|███▉      | 545000/1393783 [01:59<03:21, 4207.45 examples/s]
Tokenizing Prompts (num_proc=192):  39%|███▉      | 547000/1393783 [01:59<02:17, 6139.19 examples/s]
Tokenizing Prompts (num_proc=192):  39%|███▉      | 550000/1393783 [02:00<01:49, 7728.84 examples/s]
Tokenizing Prompts (num_proc=192):  40%|███▉      | 554000/1393783
+1:  [02:00<01:43, 8080.75 examples/s]
Tokenizing Prompts (num_proc=192):  40%|███▉      | 556000/1393783 [02:00<01:54, 7343.85 examples/s]
Tokenizing Prompts (num_proc=192):  40%|███▉      | 557000/1393783 [02:01<02:43, 5105.93 examples/s]
Tokenizing Prompts (num_proc=192):  40%|████      | 558000/1393783 [02:01<02:34, 5408.28 examples/s]
Tokenizing Prompts (num_proc=192):  40%|████      | 560000/1393783 [02:01<02:23, 5815.27 examples/s]
Tokenizing Prompts (num_proc=192):  40%|████      | 561000/1393783 [02:02<02:36, 5307.65 examples/s]
Tokenizing Prompts (num_proc=192):  40%|████      | 562000/1393783 [02:02<02:22, 5836.53 examples/s]
Tokenizing Prompts (num_proc=192):  40%|████      | 564000/1393783 [02:02<01:50, 7497.45 examples/s]
Tokenizing Prompts (num_proc=192):  41%|████      | 565000/1393783 [02:02<01:57, 7075.56 examples/s]
Tokenizing Prompts (num_proc=192):  41%|████      | 568000/1393783 [02:02<01:37, 8512.24 examples/s]
Tokenizi
+1: ng Prompts (num_proc=192):  41%|████      | 571000/1393783 [02:02<01:21, 10128.94 examples/s]
Tokenizing Prompts (num_proc=192):  41%|████      | 573000/1393783 [02:03<01:37, 8406.14 examples/s] 
Tokenizing Prompts (num_proc=192):  41%|████▏     | 577000/1393783 [02:03<01:03, 12863.11 examples/s]
Tokenizing Prompts (num_proc=192):  42%|████▏     | 579000/1393783 [02:04<02:44, 4954.92 examples/s] 
Tokenizing Prompts (num_proc=192):  42%|████▏     | 582000/1393783 [02:04<02:01, 6685.20 examples/s]
Tokenizing Prompts (num_proc=192):  42%|████▏     | 584000/1393783 [02:05<01:59, 6761.95 examples/s]
Tokenizing Prompts (num_proc=192):  42%|████▏     | 587000/1393783 [02:05<01:36, 8323.56 examples/s]
Tokenizing Prompts (num_proc=192):  42%|████▏     | 591000/1393783 [02:05<01:08, 11736.25 examples/s]
Tokenizing Prompts (num_proc=192):  43%|████▎     | 593000/1393783 [02:06<02:17, 5804.47 examples/s] 
Tokenizing Prompts (num_proc=19
+1: 2):  43%|████▎     | 595000/1393783 [02:06<01:54, 6979.81 examples/s]
Tokenizing Prompts (num_proc=192):  43%|████▎     | 597000/1393783 [02:06<01:57, 6795.99 examples/s]
Tokenizing Prompts (num_proc=192):  43%|████▎     | 599000/1393783 [02:06<01:49, 7245.60 examples/s]
Tokenizing Prompts (num_proc=192):  43%|████▎     | 601000/1393783 [02:07<01:51, 7083.14 examples/s]
Tokenizing Prompts (num_proc=192):  43%|████▎     | 603000/1393783 [02:07<01:37, 8121.89 examples/s]
Tokenizing Prompts (num_proc=192):  43%|████▎     | 605000/1393783 [02:07<01:34, 8358.33 examples/s]
Tokenizing Prompts (num_proc=192):  44%|████▎     | 607000/1393783 [02:08<02:02, 6424.58 examples/s]
Tokenizing Prompts (num_proc=192):  44%|████▎     | 608000/1393783 [02:08<02:03, 6338.37 examples/s]
Tokenizing Prompts (num_proc=192):  44%|████▎     | 609000/1393783 [02:08<02:56, 4448.45 examples/s]
Tokenizing Prompts (num_proc=192):  44%|████▍ 
+1:     | 610000/1393783 [02:08<02:37, 4982.72 examples/s]
Tokenizing Prompts (num_proc=192):  44%|████▍     | 611000/1393783 [02:09<02:41, 4858.57 examples/s]
Tokenizing Prompts (num_proc=192):  44%|████▍     | 613000/1393783 [02:09<03:17, 3957.02 examples/s]
Tokenizing Prompts (num_proc=192):  44%|████▍     | 614000/1393783 [02:10<03:15, 3987.78 examples/s]
Tokenizing Prompts (num_proc=192):  44%|████▍     | 615000/1393783 [02:10<02:53, 4499.28 examples/s]
Tokenizing Prompts (num_proc=192):  44%|████▍     | 616000/1393783 [02:10<03:20, 3888.75 examples/s]
Tokenizing Prompts (num_proc=192):  44%|████▍     | 617000/1393783 [02:10<03:53, 3324.94 examples/s]
Tokenizing Prompts (num_proc=192):  44%|████▍     | 618000/1393783 [02:11<05:33, 2326.91 examples/s]
Tokenizing Prompts (num_proc=192):  44%|████▍     | 619000/1393783 [02:11<04:39, 2768.55 examples/s]
Tokenizing Prompts (num_proc=192):  44%|████▍     | 620000/1393783 [02:
+1: 12<03:47, 3396.20 examples/s]
Tokenizing Prompts (num_proc=192):  45%|████▍     | 621000/1393783 [02:12<03:15, 3961.31 examples/s]
Tokenizing Prompts (num_proc=192):  45%|████▍     | 623000/1393783 [02:12<02:07, 6042.10 examples/s]
Tokenizing Prompts (num_proc=192):  45%|████▍     | 626000/1393783 [02:12<01:19, 9694.94 examples/s]
Tokenizing Prompts (num_proc=192):  45%|████▌     | 628000/1393783 [02:13<02:27, 5191.62 examples/s]
Tokenizing Prompts (num_proc=192):  45%|████▌     | 630000/1393783 [02:13<02:14, 5695.64 examples/s]
Tokenizing Prompts (num_proc=192):  45%|████▌     | 632000/1393783 [02:13<01:50, 6907.00 examples/s]
Tokenizing Prompts (num_proc=192):  45%|████▌     | 634000/1393783 [02:14<02:16, 5584.05 examples/s]
Tokenizing Prompts (num_proc=192):  46%|████▌     | 635000/1393783 [02:14<02:11, 5779.44 examples/s]
Tokenizing Prompts (num_proc=192):  46%|████▌     | 637000/1393783 [02:14<01:51, 6811.16 example
+1: s/s]
Tokenizing Prompts (num_proc=192):  46%|████▌     | 638000/1393783 [02:15<03:05, 4066.60 examples/s]
Tokenizing Prompts (num_proc=192):  46%|████▌     | 639000/1393783 [02:16<05:12, 2411.97 examples/s]
Tokenizing Prompts (num_proc=192):  46%|████▌     | 640000/1393783 [02:16<04:27, 2813.14 examples/s]
Tokenizing Prompts (num_proc=192):  46%|████▌     | 642000/1393783 [02:16<03:22, 3704.93 examples/s]
Tokenizing Prompts (num_proc=192):  46%|████▌     | 644000/1393783 [02:16<02:23, 5235.63 examples/s]
Tokenizing Prompts (num_proc=192):  46%|████▋     | 645000/1393783 [02:17<04:28, 2791.40 examples/s]
Tokenizing Prompts (num_proc=192):  46%|████▋     | 646000/1393783 [02:18<04:25, 2817.86 examples/s]
Tokenizing Prompts (num_proc=192):  47%|████▋     | 650000/1393783 [02:18<02:14, 5516.96 examples/s]
Tokenizing Prompts (num_proc=192):  47%|████▋     | 652000/1393783 [02:18<02:05, 5932.09 examples/s]
Tokenizing Prompts (
+1: num_proc=192):  47%|████▋     | 653000/1393783 [02:18<02:09, 5724.01 examples/s]
Tokenizing Prompts (num_proc=192):  47%|████▋     | 654000/1393783 [02:19<02:38, 4661.83 examples/s]
Tokenizing Prompts (num_proc=192):  47%|████▋     | 656000/1393783 [02:19<02:11, 5620.20 examples/s]
Tokenizing Prompts (num_proc=192):  47%|████▋     | 659000/1393783 [02:19<01:38, 7478.19 examples/s]
Tokenizing Prompts (num_proc=192):  47%|████▋     | 660000/1393783 [02:19<01:39, 7362.72 examples/s]
Tokenizing Prompts (num_proc=192):  47%|████▋     | 661000/1393783 [02:19<01:36, 7583.70 examples/s]
Tokenizing Prompts (num_proc=192):  47%|████▋     | 662000/1393783 [02:19<01:42, 7131.72 examples/s]
Tokenizing Prompts (num_proc=192):  48%|████▊     | 665000/1393783 [02:20<01:30, 8088.98 examples/s]
Tokenizing Prompts (num_proc=192):  48%|████▊     | 666000/1393783 [02:20<01:53, 6418.37 examples/s]
Tokenizing Prompts (num_proc=192):  48%|█�
+1: �██▊     | 667000/1393783 [02:20<01:57, 6177.01 examples/s]
Tokenizing Prompts (num_proc=192):  48%|████▊     | 668000/1393783 [02:20<02:00, 6034.93 examples/s]
Tokenizing Prompts (num_proc=192):  48%|████▊     | 669000/1393783 [02:21<02:02, 5930.12 examples/s]
Tokenizing Prompts (num_proc=192):  48%|████▊     | 670000/1393783 [02:21<02:02, 5927.16 examples/s]
Tokenizing Prompts (num_proc=192):  48%|████▊     | 671000/1393783 [02:22<04:16, 2815.48 examples/s]
Tokenizing Prompts (num_proc=192):  48%|████▊     | 672000/1393783 [02:22<03:42, 3244.31 examples/s]
Tokenizing Prompts (num_proc=192):  48%|████▊     | 673000/1393783 [02:22<03:25, 3508.32 examples/s]
Tokenizing Prompts (num_proc=192):  48%|████▊     | 674000/1393783 [02:22<03:45, 3192.47 examples/s]
Tokenizing Prompts (num_proc=192):  48%|████▊     | 675000/1393783 [02:23<03:08, 3814.10 examples/s]
Tokenizing Prompts (num_proc=192):  49%|████▊     | 676000/1
+1: 393783 [02:23<03:54, 3056.15 examples/s]
Tokenizing Prompts (num_proc=192):  49%|████▊     | 678000/1393783 [02:23<02:52, 4159.94 examples/s]
Tokenizing Prompts (num_proc=192):  49%|████▊     | 679000/1393783 [02:24<02:37, 4532.95 examples/s]
Tokenizing Prompts (num_proc=192):  49%|████▉     | 681000/1393783 [02:24<03:14, 3663.86 examples/s]
Tokenizing Prompts (num_proc=192):  49%|████▉     | 682000/1393783 [02:24<02:56, 4040.81 examples/s]
Tokenizing Prompts (num_proc=192):  49%|████▉     | 683000/1393783 [02:25<02:47, 4248.82 examples/s]
Tokenizing Prompts (num_proc=192):  49%|████▉     | 688000/1393783 [02:25<01:26, 8182.26 examples/s]
Tokenizing Prompts (num_proc=192):  50%|████▉     | 690000/1393783 [02:25<01:19, 8884.48 examples/s]
Tokenizing Prompts (num_proc=192):  50%|████▉     | 691000/1393783 [02:25<01:23, 8424.76 examples/s]
Tokenizing Prompts (num_proc=192):  50%|████▉     | 694000/1393783 [02:26<01:30, 7744
+1: .13 examples/s]
Tokenizing Prompts (num_proc=192):  50%|████▉     | 696000/1393783 [02:26<01:55, 6048.30 examples/s]
Tokenizing Prompts (num_proc=192):  50%|█████     | 697000/1393783 [02:26<01:54, 6090.63 examples/s]
Tokenizing Prompts (num_proc=192):  50%|█████     | 698000/1393783 [02:26<01:47, 6447.13 examples/s]
Tokenizing Prompts (num_proc=192):  50%|█████     | 699000/1393783 [02:27<01:51, 6259.13 examples/s]
Tokenizing Prompts (num_proc=192):  50%|█████     | 700000/1393783 [02:27<02:51, 4042.72 examples/s]
Tokenizing Prompts (num_proc=192):  50%|█████     | 701000/1393783 [02:27<02:36, 4418.89 examples/s]
Tokenizing Prompts (num_proc=192):  50%|█████     | 702000/1393783 [02:28<02:36, 4426.36 examples/s]
Tokenizing Prompts (num_proc=192):  51%|█████     | 704000/1393783 [02:28<02:32, 4513.88 examples/s]
Tokenizing Prompts (num_proc=192):  51%|█████     | 706000/1393783 [02:28<01:56, 5880.37 examples/s]
Tokenizin
+1: g Prompts (num_proc=192):  51%|█████     | 708000/1393783 [02:28<01:40, 6826.52 examples/s]
Tokenizing Prompts (num_proc=192):  51%|█████     | 711000/1393783 [02:29<01:39, 6840.48 examples/s]
Tokenizing Prompts (num_proc=192):  51%|█████     | 713000/1393783 [02:29<01:22, 8288.80 examples/s]
Tokenizing Prompts (num_proc=192):  51%|█████▏    | 716000/1393783 [02:29<01:13, 9216.69 examples/s]
Tokenizing Prompts (num_proc=192):  52%|█████▏    | 718000/1393783 [02:30<01:28, 7641.34 examples/s]
Tokenizing Prompts (num_proc=192):  52%|█████▏    | 720000/1393783 [02:30<01:17, 8641.44 examples/s]
Tokenizing Prompts (num_proc=192):  52%|█████▏    | 722000/1393783 [02:31<02:30, 4475.20 examples/s]
Tokenizing Prompts (num_proc=192):  52%|█████▏    | 723000/1393783 [02:31<02:29, 4498.91 examples/s]
Tokenizing Prompts (num_proc=192):  52%|█████▏    | 724000/1393783 [02:31<02:13, 5002.23 examples/s]
Tokenizing Prompts (nu
+1: m_proc=192):  52%|█████▏    | 725000/1393783 [02:31<02:07, 5245.37 examples/s]
Tokenizing Prompts (num_proc=192):  52%|█████▏    | 727000/1393783 [02:31<01:48, 6126.44 examples/s]
Tokenizing Prompts (num_proc=192):  52%|█████▏    | 729000/1393783 [02:32<01:41, 6524.73 examples/s]
Tokenizing Prompts (num_proc=192):  52%|█████▏    | 730000/1393783 [02:32<01:41, 6524.38 examples/s]
Tokenizing Prompts (num_proc=192):  53%|█████▎    | 732000/1393783 [02:32<01:33, 7101.06 examples/s]
Tokenizing Prompts (num_proc=192):  53%|█████▎    | 734000/1393783 [02:32<01:19, 8324.68 examples/s]
Tokenizing Prompts (num_proc=192):  53%|█████▎    | 735000/1393783 [02:32<01:36, 6854.05 examples/s]
Tokenizing Prompts (num_proc=192):  53%|█████▎    | 739000/1393783 [02:33<01:01, 10728.62 examples/s]
Tokenizing Prompts (num_proc=192):  53%|█████▎    | 741000/1393783 [02:33<00:59, 10902.28 examples/s]
Tokenizing Prompts (num_pro
+1: c=192):  53%|█████▎    | 745000/1393783 [02:33<00:50, 12959.90 examples/s]
Tokenizing Prompts (num_proc=192):  54%|█████▎    | 747000/1393783 [02:33<00:57, 11251.17 examples/s]
Tokenizing Prompts (num_proc=192):  54%|█████▎    | 749000/1393783 [02:34<01:46, 6066.61 examples/s] 
Tokenizing Prompts (num_proc=192):  54%|█████▍    | 750000/1393783 [02:34<01:50, 5804.12 examples/s]
Tokenizing Prompts (num_proc=192):  54%|█████▍    | 752000/1393783 [02:35<02:32, 4212.79 examples/s]
Tokenizing Prompts (num_proc=192):  54%|█████▍    | 754000/1393783 [02:35<02:20, 4540.38 examples/s]
Tokenizing Prompts (num_proc=192):  54%|█████▍    | 755000/1393783 [02:36<02:07, 5028.95 examples/s]
Tokenizing Prompts (num_proc=192):  54%|█████▍    | 756000/1393783 [02:36<02:09, 4906.86 examples/s]
Tokenizing Prompts (num_proc=192):  54%|█████▍    | 757000/1393783 [02:36<02:15, 4702.11 examples/s]
Tokenizing Prompts (num_proc=19
+1: 2):  54%|█████▍    | 759000/1393783 [02:36<01:51, 5697.15 examples/s]
Tokenizing Prompts (num_proc=192):  55%|█████▍    | 761000/1393783 [02:36<01:29, 7088.27 examples/s]
Tokenizing Prompts (num_proc=192):  55%|█████▍    | 763000/1393783 [02:37<01:13, 8626.92 examples/s]
Tokenizing Prompts (num_proc=192):  55%|█████▍    | 765000/1393783 [02:37<01:17, 8082.30 examples/s]
Tokenizing Prompts (num_proc=192):  55%|█████▍    | 766000/1393783 [02:37<02:01, 5181.55 examples/s]
Tokenizing Prompts (num_proc=192):  55%|█████▌    | 767000/1393783 [02:37<02:00, 5182.40 examples/s]
Tokenizing Prompts (num_proc=192):  55%|█████▌    | 768000/1393783 [02:38<02:00, 5174.40 examples/s]
Tokenizing Prompts (num_proc=192):  55%|█████▌    | 769000/1393783 [02:38<01:57, 5328.76 examples/s]
Tokenizing Prompts (num_proc=192):  55%|█████▌    | 770000/1393783 [02:38<01:46, 5874.77 examples/s]
Tokenizing Prompts (num_proc=192):  55
+1: %|█████▌    | 772000/1393783 [02:38<01:16, 8102.52 examples/s]
Tokenizing Prompts (num_proc=192):  56%|█████▌    | 774000/1393783 [02:38<01:13, 8437.29 examples/s]
Tokenizing Prompts (num_proc=192):  56%|█████▌    | 776000/1393783 [02:39<01:37, 6363.67 examples/s]
Tokenizing Prompts (num_proc=192):  56%|█████▌    | 777000/1393783 [02:39<01:43, 5982.46 examples/s]
Tokenizing Prompts (num_proc=192):  56%|█████▌    | 778000/1393783 [02:39<01:35, 6417.22 examples/s]
Tokenizing Prompts (num_proc=192):  56%|█████▌    | 780000/1393783 [02:40<01:57, 5206.79 examples/s]
Tokenizing Prompts (num_proc=192):  56%|█████▌    | 782000/1393783 [02:40<02:10, 4676.42 examples/s]
Tokenizing Prompts (num_proc=192):  56%|█████▌    | 783000/1393783 [02:41<02:47, 3645.72 examples/s]
Tokenizing Prompts (num_proc=192):  56%|█████▋    | 785000/1393783 [02:41<02:38, 3844.84 examples/s]
Tokenizing Prompts (num_proc=192):  56%|█�
+1: �███▋    | 786000/1393783 [02:42<03:57, 2561.43 examples/s]
Tokenizing Prompts (num_proc=192):  56%|█████▋    | 787000/1393783 [02:42<03:19, 3045.51 examples/s]
Tokenizing Prompts (num_proc=192):  57%|█████▋    | 789000/1393783 [02:43<02:57, 3412.22 examples/s]
Tokenizing Prompts (num_proc=192):  57%|█████▋    | 790000/1393783 [02:43<02:38, 3799.22 examples/s]
Tokenizing Prompts (num_proc=192):  57%|█████▋    | 793000/1393783 [02:43<01:44, 5723.16 examples/s]
Tokenizing Prompts (num_proc=192):  57%|█████▋    | 794000/1393783 [02:43<01:43, 5821.92 examples/s]
Tokenizing Prompts (num_proc=192):  57%|█████▋    | 798000/1393783 [02:44<01:35, 6226.64 examples/s]
Tokenizing Prompts (num_proc=192):  57%|█████▋    | 800000/1393783 [02:44<01:34, 6284.10 examples/s]
Tokenizing Prompts (num_proc=192):  58%|█████▊    | 803000/1393783 [02:44<01:24, 6993.93 examples/s]
Tokenizing Prompts (num_proc=192):  58%|████
+1: █▊    | 804000/1393783 [02:45<01:39, 5930.75 examples/s]
Tokenizing Prompts (num_proc=192):  58%|█████▊    | 806000/1393783 [02:45<01:32, 6346.04 examples/s]
Tokenizing Prompts (num_proc=192):  58%|█████▊    | 807000/1393783 [02:45<01:44, 5638.03 examples/s]
Tokenizing Prompts (num_proc=192):  58%|█████▊    | 808000/1393783 [02:45<01:49, 5331.12 examples/s]
Tokenizing Prompts (num_proc=192):  58%|█████▊    | 809000/1393783 [02:46<03:21, 2900.04 examples/s]
Tokenizing Prompts (num_proc=192):  58%|█████▊    | 810000/1393783 [02:46<02:48, 3468.78 examples/s]
Tokenizing Prompts (num_proc=192):  58%|█████▊    | 811000/1393783 [02:47<02:35, 3748.14 examples/s]
Tokenizing Prompts (num_proc=192):  58%|█████▊    | 813000/1393783 [02:47<02:07, 4562.09 examples/s]
Tokenizing Prompts (num_proc=192):  58%|█████▊    | 815000/1393783 [02:47<02:09, 4455.78 examples/s]
Tokenizing Prompts (num_proc=192):  59%|█████▊ 
+1:    | 816000/1393783 [02:48<02:19, 4147.21 examples/s]
Tokenizing Prompts (num_proc=192):  59%|█████▊    | 818000/1393783 [02:48<01:37, 5914.50 examples/s]
Tokenizing Prompts (num_proc=192):  59%|█████▉    | 819000/1393783 [02:48<01:30, 6362.18 examples/s]
Tokenizing Prompts (num_proc=192):  59%|█████▉    | 822000/1393783 [02:48<01:12, 7875.20 examples/s]
Tokenizing Prompts (num_proc=192):  59%|█████▉    | 823000/1393783 [02:49<01:29, 6362.90 examples/s]
Tokenizing Prompts (num_proc=192):  59%|█████▉    | 824000/1393783 [02:49<02:22, 3991.21 examples/s]
Tokenizing Prompts (num_proc=192):  59%|█████▉    | 826000/1393783 [02:49<01:40, 5677.19 examples/s]
Tokenizing Prompts (num_proc=192):  59%|█████▉    | 827000/1393783 [02:49<01:36, 5852.24 examples/s]
Tokenizing Prompts (num_proc=192):  59%|█████▉    | 828000/1393783 [02:50<01:36, 5851.44 examples/s]
Tokenizing Prompts (num_proc=192):  60%|█████▉    | 83
+1: 0000/1393783 [02:50<01:16, 7373.69 examples/s]
Tokenizing Prompts (num_proc=192):  60%|█████▉    | 832000/1393783 [02:50<01:20, 6951.21 examples/s]
Tokenizing Prompts (num_proc=192):  60%|█████▉    | 833000/1393783 [02:50<01:25, 6559.19 examples/s]
Tokenizing Prompts (num_proc=192):  60%|█████▉    | 835000/1393783 [02:50<01:11, 7850.98 examples/s]
Tokenizing Prompts (num_proc=192):  60%|█████▉    | 836000/1393783 [02:51<01:16, 7332.68 examples/s]
Tokenizing Prompts (num_proc=192):  60%|██████    | 837000/1393783 [02:51<01:33, 5956.83 examples/s]
Tokenizing Prompts (num_proc=192):  60%|██████    | 838000/1393783 [02:51<02:23, 3872.81 examples/s]
Tokenizing Prompts (num_proc=192):  60%|██████    | 839000/1393783 [02:52<02:34, 3596.59 examples/s]
Tokenizing Prompts (num_proc=192):  60%|██████    | 840000/1393783 [02:52<02:26, 3774.12 examples/s]
Tokenizing Prompts (num_proc=192):  60%|██████    | 841000/13
+1: 93783 [02:53<03:24, 2709.69 examples/s]
Tokenizing Prompts (num_proc=192):  61%|██████    | 844000/1393783 [02:53<02:29, 3683.00 examples/s]
Tokenizing Prompts (num_proc=192):  61%|██████    | 845000/1393783 [02:53<02:24, 3798.21 examples/s]
Tokenizing Prompts (num_proc=192):  61%|██████    | 847000/1393783 [02:54<01:44, 5239.10 examples/s]
Tokenizing Prompts (num_proc=192):  61%|██████    | 848000/1393783 [02:54<01:42, 5301.33 examples/s]
Tokenizing Prompts (num_proc=192):  61%|██████    | 849000/1393783 [02:54<01:39, 5474.86 examples/s]
Tokenizing Prompts (num_proc=192):  61%|██████    | 850000/1393783 [02:54<01:31, 5918.67 examples/s]
Tokenizing Prompts (num_proc=192):  61%|██████    | 851000/1393783 [02:54<02:07, 4242.97 examples/s]
Tokenizing Prompts (num_proc=192):  61%|██████    | 853000/1393783 [02:55<01:39, 5435.98 examples/s]
Tokenizing Prompts (num_proc=192):  61%|██████▏   | 854000/1393783
+1:  [02:55<02:14, 4024.58 examples/s]
Tokenizing Prompts (num_proc=192):  61%|██████▏   | 856000/1393783 [02:55<01:52, 4786.55 examples/s]
Tokenizing Prompts (num_proc=192):  61%|██████▏   | 857000/1393783 [02:56<01:44, 5114.40 examples/s]
Tokenizing Prompts (num_proc=192):  62%|██████▏   | 859000/1393783 [02:56<01:30, 5927.98 examples/s]
Tokenizing Prompts (num_proc=192):  62%|██████▏   | 861000/1393783 [02:56<01:23, 6370.47 examples/s]
Tokenizing Prompts (num_proc=192):  62%|██████▏   | 862000/1393783 [02:56<01:31, 5791.60 examples/s]
Tokenizing Prompts (num_proc=192):  62%|██████▏   | 863000/1393783 [02:56<01:29, 5962.31 examples/s]
Tokenizing Prompts (num_proc=192):  62%|██████▏   | 864000/1393783 [02:57<02:22, 3729.08 examples/s]
Tokenizing Prompts (num_proc=192):  62%|██████▏   | 867000/1393783 [02:57<01:28, 5922.81 examples/s]
Tokenizing Prompts (num_proc=192):  62%|██████▏   | 870
+1: 000/1393783 [02:57<01:03, 8279.93 examples/s]
Tokenizing Prompts (num_proc=192):  63%|██████▎   | 872000/1393783 [02:58<01:11, 7325.08 examples/s]
Tokenizing Prompts (num_proc=192):  63%|██████▎   | 874000/1393783 [02:58<01:12, 7147.75 examples/s]
Tokenizing Prompts (num_proc=192):  63%|██████▎   | 875000/1393783 [02:58<01:32, 5630.51 examples/s]
Tokenizing Prompts (num_proc=192):  63%|██████▎   | 876000/1393783 [02:59<02:46, 3115.97 examples/s]
Tokenizing Prompts (num_proc=192):  63%|██████▎   | 877000/1393783 [03:00<02:36, 3312.39 examples/s]
Tokenizing Prompts (num_proc=192):  63%|██████▎   | 878000/1393783 [03:00<02:15, 3807.49 examples/s]
Tokenizing Prompts (num_proc=192):  63%|██████▎   | 879000/1393783 [03:00<02:11, 3922.53 examples/s]
Tokenizing Prompts (num_proc=192):  63%|██████▎   | 881000/1393783 [03:00<02:06, 4067.46 examples/s]
Tokenizing Prompts (num_proc=192):  63%|██████
+1: ▎   | 882000/1393783 [03:01<02:05, 4091.70 examples/s]
Tokenizing Prompts (num_proc=192):  63%|██████▎   | 884000/1393783 [03:01<01:39, 5098.29 examples/s]
Tokenizing Prompts (num_proc=192):  64%|██████▎   | 887000/1393783 [03:01<01:07, 7496.99 examples/s]
Tokenizing Prompts (num_proc=192):  64%|██████▍   | 891000/1393783 [03:01<00:52, 9537.27 examples/s]
Tokenizing Prompts (num_proc=192):  64%|██████▍   | 895000/1393783 [03:02<00:57, 8700.34 examples/s]
Tokenizing Prompts (num_proc=192):  64%|██████▍   | 897000/1393783 [03:02<01:06, 7452.73 examples/s]
Tokenizing Prompts (num_proc=192):  65%|██████▍   | 899000/1393783 [03:03<01:09, 7140.92 examples/s]
Tokenizing Prompts (num_proc=192):  65%|██████▍   | 900000/1393783 [03:03<01:17, 6365.43 examples/s]
Tokenizing Prompts (num_proc=192):  65%|██████▍   | 902000/1393783 [03:03<01:18, 6245.41 examples/s]
Tokenizing Prompts (num_proc=192):  65%|██�
+1: ��███▍   | 904000/1393783 [03:03<01:14, 6578.82 examples/s]
Tokenizing Prompts (num_proc=192):  65%|██████▍   | 905000/1393783 [03:04<01:28, 5508.25 examples/s]
Tokenizing Prompts (num_proc=192):  65%|██████▌   | 906000/1393783 [03:04<01:21, 6014.48 examples/s]
Tokenizing Prompts (num_proc=192):  65%|██████▌   | 908000/1393783 [03:04<01:06, 7279.75 examples/s]
Tokenizing Prompts (num_proc=192):  65%|██████▌   | 909000/1393783 [03:04<01:09, 6952.61 examples/s]
Tokenizing Prompts (num_proc=192):  65%|██████▌   | 910000/1393783 [03:05<01:39, 4852.23 examples/s]
Tokenizing Prompts (num_proc=192):  65%|██████▌   | 911000/1393783 [03:05<01:50, 4382.60 examples/s]
Tokenizing Prompts (num_proc=192):  66%|██████▌   | 914000/1393783 [03:05<01:03, 7525.22 examples/s]
Tokenizing Prompts (num_proc=192):  66%|██████▌   | 916000/1393783 [03:05<01:01, 7790.24 examples/s]
Tokenizing Prompts (num_proc=192):  
+1: 66%|██████▌   | 919000/1393783 [03:05<00:44, 10599.56 examples/s]
Tokenizing Prompts (num_proc=192):  66%|██████▌   | 921000/1393783 [03:06<00:52, 9049.62 examples/s] 
Tokenizing Prompts (num_proc=192):  66%|██████▌   | 923000/1393783 [03:06<00:59, 7962.01 examples/s]
Tokenizing Prompts (num_proc=192):  66%|██████▋   | 924000/1393783 [03:07<01:30, 5187.76 examples/s]
Tokenizing Prompts (num_proc=192):  66%|██████▋   | 926000/1393783 [03:07<01:16, 6129.06 examples/s]
Tokenizing Prompts (num_proc=192):  67%|██████▋   | 927000/1393783 [03:07<01:14, 6227.34 examples/s]
Tokenizing Prompts (num_proc=192):  67%|██████▋   | 928000/1393783 [03:07<01:36, 4821.69 examples/s]
Tokenizing Prompts (num_proc=192):  67%|██████▋   | 930000/1393783 [03:07<01:12, 6432.56 examples/s]
Tokenizing Prompts (num_proc=192):  67%|██████▋   | 931000/1393783 [03:08<01:21, 5700.13 examples/s]
Tokenizing Prompts (num
+1: _proc=192):  67%|██████▋   | 932000/1393783 [03:08<01:22, 5581.60 examples/s]
Tokenizing Prompts (num_proc=192):  67%|██████▋   | 934000/1393783 [03:08<00:59, 7786.09 examples/s]
Tokenizing Prompts (num_proc=192):  67%|██████▋   | 936000/1393783 [03:08<00:53, 8496.05 examples/s]
Tokenizing Prompts (num_proc=192):  67%|██████▋   | 938000/1393783 [03:08<00:58, 7856.32 examples/s]
Tokenizing Prompts (num_proc=192):  67%|██████▋   | 939000/1393783 [03:09<01:16, 5962.58 examples/s]
Tokenizing Prompts (num_proc=192):  67%|██████▋   | 940000/1393783 [03:09<01:21, 5547.52 examples/s]
Tokenizing Prompts (num_proc=192):  68%|██████▊   | 941000/1393783 [03:09<01:19, 5697.89 examples/s]
Tokenizing Prompts (num_proc=192):  68%|██████▊   | 942000/1393783 [03:09<01:30, 4965.17 examples/s]
Tokenizing Prompts (num_proc=192):  68%|██████▊   | 943000/1393783 [03:10<01:40, 4465.31 examples/s]
Tokenizing P
+1: rompts (num_proc=192):  68%|██████▊   | 944000/1393783 [03:10<01:44, 4315.55 examples/s]
Tokenizing Prompts (num_proc=192):  68%|██████▊   | 945000/1393783 [03:10<01:37, 4619.67 examples/s]
Tokenizing Prompts (num_proc=192):  68%|██████▊   | 947000/1393783 [03:10<01:19, 5589.61 examples/s]
Tokenizing Prompts (num_proc=192):  68%|██████▊   | 948000/1393783 [03:11<01:16, 5794.62 examples/s]
Tokenizing Prompts (num_proc=192):  68%|██████▊   | 950000/1393783 [03:11<01:01, 7263.01 examples/s]
Tokenizing Prompts (num_proc=192):  68%|██████▊   | 951000/1393783 [03:11<01:11, 6165.10 examples/s]
Tokenizing Prompts (num_proc=192):  68%|██████▊   | 952000/1393783 [03:11<01:13, 5977.33 examples/s]
Tokenizing Prompts (num_proc=192):  68%|██████▊   | 953000/1393783 [03:11<01:11, 6178.21 examples/s]
Tokenizing Prompts (num_proc=192):  69%|██████▊   | 955000/1393783 [03:12<01:00, 7234.80 examples/s]
T
+1: okenizing Prompts (num_proc=192):  69%|██████▊   | 956000/1393783 [03:12<01:09, 6290.80 examples/s]
Tokenizing Prompts (num_proc=192):  69%|██████▊   | 957000/1393783 [03:12<01:28, 4916.69 examples/s]
Tokenizing Prompts (num_proc=192):  69%|██████▊   | 958000/1393783 [03:12<01:37, 4458.79 examples/s]
Tokenizing Prompts (num_proc=192):  69%|██████▉   | 960000/1393783 [03:13<01:08, 6341.50 examples/s]
Tokenizing Prompts (num_proc=192):  69%|██████▉   | 961000/1393783 [03:13<01:05, 6597.54 examples/s]
Tokenizing Prompts (num_proc=192):  69%|██████▉   | 962000/1393783 [03:13<01:03, 6756.71 examples/s]
Tokenizing Prompts (num_proc=192):  69%|██████▉   | 963000/1393783 [03:13<01:12, 5929.17 examples/s]
Tokenizing Prompts (num_proc=192):  69%|██████▉   | 965000/1393783 [03:14<01:22, 5180.35 examples/s]
Tokenizing Prompts (num_proc=192):  69%|██████▉   | 967000/1393783 [03:14<01:03, 6715.55 ex
+1: amples/s]
Tokenizing Prompts (num_proc=192):  69%|██████▉   | 968000/1393783 [03:14<01:25, 4970.23 examples/s]
Tokenizing Prompts (num_proc=192):  70%|██████▉   | 970000/1393783 [03:14<01:03, 6654.97 examples/s]
Tokenizing Prompts (num_proc=192):  70%|██████▉   | 971000/1393783 [03:14<01:13, 5730.40 examples/s]
Tokenizing Prompts (num_proc=192):  70%|██████▉   | 973000/1393783 [03:15<01:10, 5998.38 examples/s]
Tokenizing Prompts (num_proc=192):  70%|██████▉   | 974000/1393783 [03:15<01:42, 4093.85 examples/s]
Tokenizing Prompts (num_proc=192):  70%|██████▉   | 975000/1393783 [03:15<01:29, 4653.94 examples/s]
Tokenizing Prompts (num_proc=192):  70%|███████   | 977000/1393783 [03:16<01:07, 6174.36 examples/s]
Tokenizing Prompts (num_proc=192):  70%|███████   | 979000/1393783 [03:16<00:54, 7582.51 examples/s]
Tokenizing Prompts (num_proc=192):  70%|███████   | 981000/1393783 [03:16<01:03,
+1:  6491.31 examples/s]
Tokenizing Prompts (num_proc=192):  71%|███████   | 983000/1393783 [03:16<00:56, 7279.23 examples/s]
Tokenizing Prompts (num_proc=192):  71%|███████   | 984000/1393783 [03:17<01:06, 6192.28 examples/s]
Tokenizing Prompts (num_proc=192):  71%|███████   | 985000/1393783 [03:17<01:29, 4561.53 examples/s]
Tokenizing Prompts (num_proc=192):  71%|███████   | 987000/1393783 [03:17<01:24, 4825.58 examples/s]
Tokenizing Prompts (num_proc=192):  71%|███████   | 989000/1393783 [03:18<01:26, 4676.17 examples/s]
Tokenizing Prompts (num_proc=192):  71%|███████   | 991000/1393783 [03:19<01:53, 3542.66 examples/s]
Tokenizing Prompts (num_proc=192):  71%|███████   | 992000/1393783 [03:19<01:41, 3956.49 examples/s]
Tokenizing Prompts (num_proc=192):  71%|███████   | 993000/1393783 [03:19<01:31, 4378.05 examples/s]
Tokenizing Prompts (num_proc=192):  71%|███████▏  | 994000/1393783 
+1: [03:19<01:46, 3742.49 examples/s]
Tokenizing Prompts (num_proc=192):  71%|███████▏  | 996000/1393783 [03:20<01:22, 4849.06 examples/s]
Tokenizing Prompts (num_proc=192):  72%|███████▏  | 997000/1393783 [03:20<01:16, 5174.95 examples/s]
Tokenizing Prompts (num_proc=192):  72%|███████▏  | 998000/1393783 [03:20<01:11, 5568.09 examples/s]
Tokenizing Prompts (num_proc=192):  72%|███████▏  | 999000/1393783 [03:20<01:06, 5895.54 examples/s]
Tokenizing Prompts (num_proc=192):  72%|███████▏  | 1000000/1393783 [03:20<01:10, 5610.45 examples/s]
Tokenizing Prompts (num_proc=192):  72%|███████▏  | 1003000/1393783 [03:20<00:41, 9501.53 examples/s]
Tokenizing Prompts (num_proc=192):  72%|███████▏  | 1005000/1393783 [03:23<02:51, 2264.46 examples/s]
Tokenizing Prompts (num_proc=192):  72%|███████▏  | 1006000/1393783 [03:23<02:47, 2309.82 examples/s]
Tokenizing Prompts (num_proc=192):  72%|███�
+1: ��███▏  | 1007000/1393783 [03:23<02:20, 2751.80 examples/s]
Tokenizing Prompts (num_proc=192):  72%|███████▏  | 1009000/1393783 [03:23<01:33, 4098.07 examples/s]
Tokenizing Prompts (num_proc=192):  73%|███████▎  | 1011000/1393783 [03:25<02:31, 2525.42 examples/s]
Tokenizing Prompts (num_proc=192):  73%|███████▎  | 1014000/1393783 [03:25<01:31, 4148.56 examples/s]
Tokenizing Prompts (num_proc=192):  73%|███████▎  | 1016000/1393783 [03:25<01:10, 5387.71 examples/s]
Tokenizing Prompts (num_proc=192):  73%|███████▎  | 1018000/1393783 [03:25<01:22, 4561.40 examples/s]
Tokenizing Prompts (num_proc=192):  73%|███████▎  | 1020000/1393783 [03:26<01:07, 5567.10 examples/s]
Tokenizing Prompts (num_proc=192):  73%|███████▎  | 1022000/1393783 [03:26<01:25, 4340.86 examples/s]
Tokenizing Prompts (num_proc=192):  73%|███████▎  | 1024000/1393783 [03:26<01:09, 5283.60 examples/s]
Tokenizing P
+1: rompts (num_proc=192):  74%|███████▎  | 1025000/1393783 [03:27<01:05, 5620.90 examples/s]
Tokenizing Prompts (num_proc=192):  74%|███████▎  | 1027000/1393783 [03:27<00:52, 7027.85 examples/s]
Tokenizing Prompts (num_proc=192):  74%|███████▍  | 1029260/1393783 [03:27<00:43, 8305.15 examples/s]
Tokenizing Prompts (num_proc=192):  74%|███████▍  | 1031260/1393783 [03:27<00:41, 8660.17 examples/s]
Tokenizing Prompts (num_proc=192):  74%|███████▍  | 1033260/1393783 [03:27<00:39, 9106.72 examples/s]
Tokenizing Prompts (num_proc=192):  74%|███████▍  | 1035260/1393783 [03:28<00:45, 7930.39 examples/s]
Tokenizing Prompts (num_proc=192):  74%|███████▍  | 1036260/1393783 [03:28<00:57, 6184.10 examples/s]
Tokenizing Prompts (num_proc=192):  74%|███████▍  | 1038260/1393783 [03:28<01:03, 5641.48 examples/s]
Tokenizing Prompts (num_proc=192):  75%|███████▍  | 1039520/1393783 [03:29<0
+1: 1:13, 4806.28 examples/s]
Tokenizing Prompts (num_proc=192):  75%|███████▍  | 1040520/1393783 [03:29<01:21, 4330.83 examples/s]
Tokenizing Prompts (num_proc=192):  75%|███████▍  | 1041520/1393783 [03:29<01:14, 4736.68 examples/s]
Tokenizing Prompts (num_proc=192):  75%|███��███▍  | 1043520/1393783 [03:29<00:52, 6659.20 examples/s]
Tokenizing Prompts (num_proc=192):  75%|███████▍  | 1044520/1393783 [03:29<00:52, 6703.88 examples/s]
Tokenizing Prompts (num_proc=192):  75%|███████▌  | 1045520/1393783 [03:30<01:00, 5754.96 examples/s]
Tokenizing Prompts (num_proc=192):  75%|███████▌  | 1047520/1393783 [03:30<00:44, 7844.59 examples/s]
Tokenizing Prompts (num_proc=192):  75%|███████▌  | 1048520/1393783 [03:30<00:50, 6781.43 examples/s]
Tokenizing Prompts (num_proc=192):  75%|███████▌  | 1049520/1393783 [03:30<01:14, 4602.30 examples/s]
Tokenizing Prompts (num_proc=192):  75%|████�
+1: �██▌  | 1050520/1393783 [03:31<01:04, 5321.34 examples/s]
Tokenizing Prompts (num_proc=192):  75%|███████▌  | 1051520/1393783 [03:31<01:01, 5584.87 examples/s]
Tokenizing Prompts (num_proc=192):  76%|███████▌  | 1052520/1393783 [03:31<00:58, 5789.47 examples/s]
Tokenizing Prompts (num_proc=192):  76%|███████▌  | 1053520/1393783 [03:31<01:03, 5318.54 examples/s]
Tokenizing Prompts (num_proc=192):  76%|███████▌  | 1055520/1393783 [03:31<00:48, 6911.47 examples/s]
Tokenizing Prompts (num_proc=192):  76%|███████▌  | 1056520/1393783 [03:31<00:50, 6615.98 examples/s]
Tokenizing Prompts (num_proc=192):  76%|███████▌  | 1057520/1393783 [03:32<00:54, 6191.57 examples/s]
Tokenizing Prompts (num_proc=192):  76%|███████▌  | 1058520/1393783 [03:32<00:49, 6758.39 examples/s]
Tokenizing Prompts (num_proc=192):  76%|███████▌  | 1060520/1393783 [03:32<00:40, 8259.00 examples/s]
Tokenizing Promp
+1: ts (num_proc=192):  76%|███████▋  | 1063520/1393783 [03:32<00:28, 11612.21 examples/s]
Tokenizing Prompts (num_proc=192):  76%|███████▋  | 1065520/1393783 [03:32<00:34, 9606.54 examples/s] 
Tokenizing Prompts (num_proc=192):  77%|███████▋  | 1067520/1393783 [03:33<00:42, 7671.08 examples/s]
Tokenizing Prompts (num_proc=192):  77%|███████▋  | 1068520/1393783 [03:33<00:40, 7943.54 examples/s]
Tokenizing Prompts (num_proc=192):  77%|███████▋  | 1069520/1393783 [03:34<01:20, 4051.03 examples/s]
Tokenizing Prompts (num_proc=192):  77%|███████▋  | 1070520/1393783 [03:34<01:22, 3911.33 examples/s]
Tokenizing Prompts (num_proc=192):  77%|███████▋  | 1071520/1393783 [03:34<01:11, 4532.35 examples/s]
Tokenizing Prompts (num_proc=192):  77%|███████▋  | 1072520/1393783 [03:34<01:05, 4878.26 examples/s]
Tokenizing Prompts (num_proc=192):  77%|███████▋  | 1073520/1393783 [03:34<01:
+1: 10, 4522.88 examples/s]
Tokenizing Prompts (num_proc=192):  77%|███████▋  | 1074520/1393783 [03:35<01:05, 4847.26 examples/s]
Tokenizing Prompts (num_proc=192):  77%|███████▋  | 1075520/1393783 [03:35<00:57, 5494.25 examples/s]
Tokenizing Prompts (num_proc=192):  77%|███████▋  | 1077040/1393783 [03:35<00:46, 6762.74 examples/s]
Tokenizing Prompts (num_proc=192):  77%|███████▋  | 1078300/1393783 [03:35<00:53, 5944.47 examples/s]
Tokenizing Prompts (num_proc=192):  78%|███████▊  | 1080300/1393783 [03:35<00:52, 6015.49 examples/s]
Tokenizing Prompts (num_proc=192):  78%|███████▊  | 1081300/1393783 [03:36<00:55, 5580.93 examples/s]
Tokenizing Prompts (num_proc=192):  78%|███████▊  | 1084300/1393783 [03:36<00:44, 7018.10 examples/s]
Tokenizing Prompts (num_proc=192):  78%|███████▊  | 1086300/1393783 [03:36<00:38, 7933.01 examples/s]
Tokenizing Prompts (num_proc=192):  78%|█████�
+1: ��█▊  | 1089300/1393783 [03:36<00:28, 10573.99 examples/s]
Tokenizing Prompts (num_proc=192):  78%|███████▊  | 1090820/1393783 [03:36<00:29, 10293.73 examples/s]
Tokenizing Prompts (num_proc=192):  78%|███████▊  | 1092820/1393783 [03:37<00:27, 10958.29 examples/s]
Tokenizing Prompts (num_proc=192):  79%|███████▊  | 1094820/1393783 [03:37<00:32, 9305.51 examples/s] 
Tokenizing Prompts (num_proc=192):  79%|███████▊  | 1096820/1393783 [03:37<00:38, 7651.56 examples/s]
Tokenizing Prompts (num_proc=192):  79%|███████▉  | 1098080/1393783 [03:38<00:49, 5979.07 examples/s]
Tokenizing Prompts (num_proc=192):  79%|███████▉  | 1101080/1393783 [03:38<00:37, 7843.38 examples/s]
Tokenizing Prompts (num_proc=192):  79%|███████▉  | 1102339/1393783 [03:38<00:41, 7085.12 examples/s]
Tokenizing Prompts (num_proc=192):  79%|███████▉  | 1104339/1393783 [03:38<00:35, 8116.95 examples/s]
Tokenizing Pro
+1: mpts (num_proc=192):  79%|███████▉  | 1105599/1393783 [03:39<00:40, 7067.05 examples/s]
Tokenizing Prompts (num_proc=192):  79%|███████▉  | 1106599/1393783 [03:39<00:51, 5584.12 examples/s]
Tokenizing Prompts (num_proc=192):  79%|███████▉  | 1107599/1393783 [03:39<00:46, 6122.20 examples/s]
Tokenizing Prompts (num_proc=192):  80%|███████▉  | 1108599/1393783 [03:39<00:46, 6082.49 examples/s]
Tokenizing Prompts (num_proc=192):  80%|███████▉  | 1109859/1393783 [03:39<00:46, 6057.38 examples/s]
Tokenizing Prompts (num_proc=192):  80%|███████▉  | 1110859/1393783 [03:40<00:51, 5483.17 examples/s]
Tokenizing Prompts (num_proc=192):  80%|███████▉  | 1111638/1393783 [03:40<00:56, 4985.09 examples/s]
Tokenizing Prompts (num_proc=192):  80%|███████▉  | 1112638/1393783 [03:40<00:52, 5402.18 examples/s]
Tokenizing Prompts (num_proc=192):  80%|███████▉  | 1113898/1393783 [03:40<00:
+1: 47, 5896.49 examples/s]
Tokenizing Prompts (num_proc=192):  80%|███████▉  | 1114898/1393783 [03:40<00:43, 6391.36 examples/s]
Tokenizing Prompts (num_proc=192):  80%|████████  | 1116898/1393783 [03:41<00:49, 5648.64 examples/s]
Tokenizing Prompts (num_proc=192):  80%|████████  | 1117898/1393783 [03:41<00:44, 6190.01 examples/s]
Tokenizing Prompts (num_proc=192):  80%|████████  | 1120418/1393783 [03:41<00:47, 5721.77 examples/s]
Tokenizing Prompts (num_proc=192):  81%|████████  | 1122678/1393783 [03:41<00:36, 7359.76 examples/s]
Tokenizing Prompts (num_proc=192):  81%|████████  | 1123678/1393783 [03:41<00:35, 7621.78 examples/s]
Tokenizing Prompts (num_proc=192):  81%|████████  | 1124937/1393783 [03:42<00:31, 8452.10 examples/s]
Tokenizing Prompts (num_proc=192):  81%|████████  | 1125937/1393783 [03:42<00:37, 7089.83 examples/s]
Tokenizing Prompts (num_proc=192):  81%|█████�
+1: ��██  | 1126937/1393783 [03:42<00:39, 6779.65 examples/s]
Tokenizing Prompts (num_proc=192):  81%|████████  | 1128937/1393783 [03:42<00:29, 9024.44 examples/s]
Tokenizing Prompts (num_proc=192):  81%|████████  | 1130937/1393783 [03:42<00:31, 8358.54 examples/s]
Tokenizing Prompts (num_proc=192):  81%|████████▏ | 1133937/1393783 [03:43<00:31, 8151.45 examples/s]
Tokenizing Prompts (num_proc=192):  81%|████████▏ | 1135197/1393783 [03:43<00:34, 7536.59 examples/s]
Tokenizing Prompts (num_proc=192):  82%|████████▏ | 1137197/1393783 [03:43<00:31, 8193.94 examples/s]
Tokenizing Prompts (num_proc=192):  82%|████████▏ | 1138197/1393783 [03:43<00:36, 7091.27 examples/s]
Tokenizing Prompts (num_proc=192):  82%|████████▏ | 1139197/1393783 [03:43<00:34, 7368.73 examples/s]
Tokenizing Prompts (num_proc=192):  82%|████████▏ | 1140457/1393783 [03:44<00:34, 7341.52 examples/s]
Tokeni
+1: zing Prompts (num_proc=192):  82%|████████▏ | 1141236/1393783 [03:44<00:53, 4749.08 examples/s]
Tokenizing Prompts (num_proc=192):  82%|████████▏ | 1142496/1393783 [03:44<00:54, 4629.55 examples/s]
Tokenizing Prompts (num_proc=192):  82%|████████▏ | 1144756/1393783 [03:45<00:52, 4742.04 examples/s]
Tokenizing Prompts (num_proc=192):  82%|████████▏ | 1146756/1393783 [03:45<00:41, 5893.96 examples/s]
Tokenizing Prompts (num_proc=192):  82%|████████▏ | 1147756/1393783 [03:45<00:41, 5996.44 examples/s]
Tokenizing Prompts (num_proc=192):  82%|████████▏ | 1149756/1393783 [03:45<00:35, 6971.66 examples/s]
Tokenizing Prompts (num_proc=192):  83%|████████▎ | 1153016/1393783 [03:45<00:22, 10627.94 examples/s]
Tokenizing Prompts (num_proc=192):  83%|████████▎ | 1154534/1393783 [03:46<00:24, 9635.83 examples/s] 
Tokenizing Prompts (num_proc=192):  83%|████████▎ 
+1: | 1156534/1393783 [03:46<00:29, 8133.23 examples/s]
Tokenizing Prompts (num_proc=192):  83%|████████▎ | 1157794/1393783 [03:46<00:31, 7587.24 examples/s]
Tokenizing Prompts (num_proc=192):  83%|████████▎ | 1161794/1393783 [03:46<00:21, 10840.41 examples/s]
Tokenizing Prompts (num_proc=192):  83%|████████▎ | 1163794/1393783 [03:47<00:19, 11528.43 examples/s]
Tokenizing Prompts (num_proc=192):  84%|████████▎ | 1165054/1393783 [03:47<00:31, 7181.14 examples/s] 
Tokenizing Prompts (num_proc=192):  84%|████████▎ | 1166314/1393783 [03:47<00:35, 6471.12 examples/s]
Tokenizing Prompts (num_proc=192):  84%|████████▍ | 1167314/1393783 [03:47<00:33, 6814.49 examples/s]
Tokenizing Prompts (num_proc=192):  84%|████████▍ | 1168573/1393783 [03:48<00:31, 7106.03 examples/s]
Tokenizing Prompts (num_proc=192):  84%|████████▍ | 1169573/1393783 [03:48<00:31, 7036.64 examples/s]
Tokenizin
+1: g Prompts (num_proc=192):  84%|████████▍ | 1170573/1393783 [03:48<00:31, 7119.08 examples/s]
Tokenizing Prompts (num_proc=192):  84%|████████▍ | 1171832/1393783 [03:48<00:39, 5558.37 examples/s]
Tokenizing Prompts (num_proc=192):  84%|████████▍ | 1172832/1393783 [03:48<00:36, 6007.88 examples/s]
Tokenizing Prompts (num_proc=192):  85%|████████▍ | 1177832/1393783 [03:49<00:29, 7256.18 examples/s]
Tokenizing Prompts (num_proc=192):  85%|████████▍ | 1181092/1393783 [03:49<00:29, 7114.70 examples/s]
Tokenizing Prompts (num_proc=192):  85%|████████▍ | 1182092/1393783 [03:50<00:29, 7124.06 examples/s]
Tokenizing Prompts (num_proc=192):  85%|████████▍ | 1183610/1393783 [03:50<00:26, 7859.08 examples/s]
Tokenizing Prompts (num_proc=192):  85%|████████▌ | 1184869/1393783 [03:50<00:25, 8163.44 examples/s]
Tokenizing Prompts (num_proc=192):  85%|████████▌ | 118
+1: 6648/1393783 [03:50<00:34, 6064.41 examples/s]
Tokenizing Prompts (num_proc=192):  85%|████████▌ | 1187907/1393783 [03:51<00:37, 5489.75 examples/s]
Tokenizing Prompts (num_proc=192):  85%|████████▌ | 1190907/1393783 [03:51<00:26, 7550.78 examples/s]
Tokenizing Prompts (num_proc=192):  86%|████████▌ | 1191907/1393783 [03:51<00:25, 7847.31 examples/s]
Tokenizing Prompts (num_proc=192):  86%|████████▌ | 1192907/1393783 [03:51<00:31, 6292.92 examples/s]
Tokenizing Prompts (num_proc=192):  86%|████████▌ | 1194425/1393783 [03:51<00:35, 5643.78 examples/s]
Tokenizing Prompts (num_proc=192):  86%|████████▌ | 1195944/1393783 [03:52<00:54, 3597.09 examples/s]
Tokenizing Prompts (num_proc=192):  86%|████████▌ | 1197944/1393783 [03:53<00:42, 4563.70 examples/s]
Tokenizing Prompts (num_proc=192):  86%|████████▌ | 1199944/1393783 [03:53<00:37, 5171.57 examples/s]
Tokenizing Prompt
+1: s (num_proc=192):  86%|████████▌ | 1201944/1393783 [03:53<00:30, 6285.63 examples/s]
Tokenizing Prompts (num_proc=192):  86%|████████▋ | 1203463/1393783 [03:53<00:33, 5737.50 examples/s]
Tokenizing Prompts (num_proc=192):  86%|████████▋ | 1204723/1393783 [03:53<00:29, 6304.92 examples/s]
Tokenizing Prompts (num_proc=192):  86%|████████▋ | 1205501/1393783 [03:54<00:30, 6133.06 examples/s]
Tokenizing Prompts (num_proc=192):  87%|████████▋ | 1207278/1393783 [03:54<00:24, 7683.63 examples/s]
Tokenizing Prompts (num_proc=192):  87%|████████▋ | 1208797/1393783 [03:54<00:25, 7129.84 examples/s]
Tokenizing Prompts (num_proc=192):  87%|████████▋ | 1209797/1393783 [03:54<00:27, 6802.40 examples/s]
Tokenizing Prompts (num_proc=192):  87%|████████▋ | 1211797/1393783 [03:54<00:25, 7073.16 examples/s]
Tokenizing Prompts (num_proc=192):  87%|████████▋ | 1212797/139
+1: 3783 [03:55<00:28, 6369.63 examples/s]
Tokenizing Prompts (num_proc=192):  87%|████████▋ | 1213575/1393783 [03:55<00:53, 3389.90 examples/s]
Tokenizing Prompts (num_proc=192):  87%|████████▋ | 1214575/1393783 [03:55<00:47, 3747.05 examples/s]
Tokenizing Prompts (num_proc=192):  87%|████████▋ | 1215834/1393783 [03:56<00:37, 4803.06 examples/s]
Tokenizing Prompts (num_proc=192):  87%|████████▋ | 1216834/1393783 [03:56<00:32, 5417.60 examples/s]
Tokenizing Prompts (num_proc=192):  88%|████████▊ | 1220093/1393783 [03:56<00:18, 9513.92 examples/s]
Tokenizing Prompts (num_proc=192):  88%|████████▊ | 1221352/1393783 [03:56<00:21, 7840.56 examples/s]
Tokenizing Prompts (num_proc=192):  88%|████████▊ | 1223352/1393783 [03:56<00:25, 6706.77 examples/s]
Tokenizing Prompts (num_proc=192):  88%|████████▊ | 1224352/1393783 [03:57<00:24, 7008.14 examples/s]
Tokenizing Prompts (num_p
+1: roc=192):  88%|████████▊ | 1225389/1393783 [03:57<00:26, 6315.57 examples/s]
Tokenizing Prompts (num_proc=192):  88%|████████▊ | 1226166/1393783 [03:57<00:30, 5533.72 examples/s]
Tokenizing Prompts (num_proc=192):  88%|████████▊ | 1227684/1393783 [03:57<00:31, 5199.52 examples/s]
Tokenizing Prompts (num_proc=192):  88%|████████▊ | 1232684/1393783 [03:58<00:18, 8596.91 examples/s]
Tokenizing Prompts (num_proc=192):  89%|████████▊ | 1233943/1393783 [03:58<00:17, 8976.40 examples/s]
Tokenizing Prompts (num_proc=192):  89%|████████▊ | 1236203/1393783 [03:58<00:14, 10627.04 examples/s]
Tokenizing Prompts (num_proc=192):  89%|████████▉ | 1238203/1393783 [03:58<00:14, 11072.95 examples/s]
Tokenizing Prompts (num_proc=192):  89%|████████▉ | 1239462/1393783 [03:58<00:14, 10879.38 examples/s]
Tokenizing Prompts (num_proc=192):  89%|████████▉ | 1241721/1393783 
+1: [03:58<00:12, 12334.46 examples/s]
Tokenizing Prompts (num_proc=192):  89%|████████▉ | 1243239/1393783 [03:59<00:28, 5376.27 examples/s] 
Tokenizing Prompts (num_proc=192):  89%|████████▉ | 1244239/1393783 [03:59<00:29, 5124.40 examples/s]
Tokenizing Prompts (num_proc=192):  89%|████████▉ | 1245757/1393783 [04:00<00:43, 3427.32 examples/s]
Tokenizing Prompts (num_proc=192):  89%|████████▉ | 1247275/1393783 [04:00<00:37, 3859.59 examples/s]
Tokenizing Prompts (num_proc=192):  90%|████████▉ | 1248275/1393783 [04:01<00:38, 3777.76 examples/s]
Tokenizing Prompts (num_proc=192):  90%|████████▉ | 1250534/1393783 [04:01<00:25, 5711.39 examples/s]
Tokenizing Prompts (num_proc=192):  90%|████████▉ | 1252534/1393783 [04:01<00:22, 6290.47 examples/s]
Tokenizing Prompts (num_proc=192):  90%|█████████ | 1254793/1393783 [04:01<00:19, 7249.62 examples/s]
Tokenizing Prompts (num_proc
+1: =192):  90%|█████████ | 1256052/1393783 [04:01<00:18, 7355.96 examples/s]
Tokenizing Prompts (num_proc=192):  90%|█████████ | 1257311/1393783 [04:02<00:16, 8091.60 examples/s]
Tokenizing Prompts (num_proc=192):  90%|█████████ | 1260570/1393783 [04:02<00:14, 9494.36 examples/s]
Tokenizing Prompts (num_proc=192):  91%|█████████ | 1262088/1393783 [04:02<00:16, 7773.55 examples/s]
Tokenizing Prompts (num_proc=192):  91%|█████████ | 1263866/1393783 [04:02<00:19, 6732.36 examples/s]
Tokenizing Prompts (num_proc=192):  91%|█████████ | 1265125/1393783 [04:03<00:18, 6832.06 examples/s]
Tokenizing Prompts (num_proc=192):  91%|█████████ | 1267384/1393783 [04:03<00:15, 8333.28 examples/s]
Tokenizing Prompts (num_proc=192):  91%|█████████ | 1268384/1393783 [04:03<00:14, 8371.01 examples/s]
Tokenizing Prompts (num_proc=192):  91%|█████████ | 1269643/1393783 [04:04
+1: <00:28, 4419.29 examples/s]
Tokenizing Prompts (num_proc=192):  91%|█████████ | 1270643/1393783 [04:04<00:25, 4755.60 examples/s]
Tokenizing Prompts (num_proc=192):  91%|█████████▏| 1271902/1393783 [04:04<00:25, 4778.94 examples/s]
Tokenizing Prompts (num_proc=192):  91%|█████████▏| 1272902/1393783 [04:04<00:23, 5056.36 examples/s]
Tokenizing Prompts (num_proc=192):  91%|█████████▏| 1274902/1393783 [04:04<00:19, 6235.86 examples/s]
Tokenizing Prompts (num_proc=192):  92%|█████████▏| 1278161/1393783 [04:05<00:12, 9373.59 examples/s]
Tokenizing Prompts (num_proc=192):  92%|█████████▏| 1281161/1393783 [04:05<00:10, 10414.89 examples/s]
Tokenizing Prompts (num_proc=192):  92%|█████████▏| 1282420/1393783 [04:05<00:14, 7564.33 examples/s] 
Tokenizing Prompts (num_proc=192):  92%|█████████▏| 1284198/1393783 [04:06<00:19, 5635.70 examples/s]
Tokenizing Prompts (
+1: num_proc=192):  92%|█████████▏| 1286716/1393783 [04:06<00:14, 7414.72 examples/s]
Tokenizing Prompts (num_proc=192):  92%|█████████▏| 1288234/1393783 [04:06<00:12, 8301.97 examples/s]
Tokenizing Prompts (num_proc=192):  93%|█████████▎| 1291234/1393783 [04:06<00:10, 9361.01 examples/s]
Tokenizing Prompts (num_proc=192):  93%|█████████▎| 1293234/1393783 [04:06<00:09, 10636.21 examples/s]
Tokenizing Prompts (num_proc=192):  93%|█████████▎| 1295493/1393783 [04:07<00:15, 6172.46 examples/s] 
Tokenizing Prompts (num_proc=192):  93%|█████████▎| 1300011/1393783 [04:07<00:09, 9889.93 examples/s]
Tokenizing Prompts (num_proc=192):  93%|█████████▎| 1303011/1393783 [04:07<00:07, 11546.12 examples/s]
Tokenizing Prompts (num_proc=192):  94%|█████████▎| 1305011/1393783 [04:07<00:07, 12409.48 examples/s]
Tokenizing Prompts (num_proc=192):  94%|████████
+1: █▍| 1307271/1393783 [04:08<00:13, 6488.05 examples/s] 
Tokenizing Prompts (num_proc=192):  94%|█████████▍| 1309530/1393783 [04:09<00:11, 7034.35 examples/s]
Tokenizing Prompts (num_proc=192):  94%|█████████▍| 1311307/1393783 [04:09<00:16, 5088.86 examples/s]
Tokenizing Prompts (num_proc=192):  94%|█████████▍| 1313307/1393783 [04:10<00:19, 4158.98 examples/s]
Tokenizing Prompts (num_proc=192):  94%|█████████▍| 1314567/1393783 [04:10<00:16, 4720.73 examples/s]
Tokenizing Prompts (num_proc=192):  94%|█████████▍| 1316086/1393783 [04:10<00:14, 5319.79 examples/s]
Tokenizing Prompts (num_proc=192):  94%|█████████▍| 1317122/1393783 [04:11<00:20, 3683.01 examples/s]
Tokenizing Prompts (num_proc=192):  95%|█████████▍| 1318640/1393783 [04:11<00:22, 3346.14 examples/s]
Tokenizing Prompts (num_proc=192):  95%|█████████▍| 1319158/1393783 [04:12<00:22, 3353.81 e
+1: xamples/s]
Tokenizing Prompts (num_proc=192):  95%|█████████▍| 1322158/1393783 [04:12<00:12, 5904.74 examples/s]
Tokenizing Prompts (num_proc=192):  95%|█████████▌| 1324676/1393783 [04:12<00:08, 7747.00 examples/s]
Tokenizing Prompts (num_proc=192):  95%|█████████▌| 1326195/1393783 [04:13<00:14, 4566.20 examples/s]
Tokenizing Prompts (num_proc=192):  95%|█████████▌| 1328195/1393783 [04:13<00:11, 5928.61 examples/s]
Tokenizing Prompts (num_proc=192):  96%|█████████▌| 1331195/1393783 [04:13<00:08, 7417.22 examples/s]
Tokenizing Prompts (num_proc=192):  96%|█████████▌| 1332713/1393783 [04:13<00:07, 8043.46 examples/s]
Tokenizing Prompts (num_proc=192):  96%|█████████▌| 1334973/1393783 [04:14<00:11, 5152.81 examples/s]
Tokenizing Prompts (num_proc=192):  96%|█████████▌| 1336493/1393783 [04:15<00:15, 3654.66 examples/s]
Tokenizing Prompts (num_proc=192):  9
+1: 6%|█████████▌| 1337272/1393783 [04:15<00:17, 3149.60 examples/s]
Tokenizing Prompts (num_proc=192):  96%|█████████▌| 1339272/1393783 [04:15<00:14, 3737.00 examples/s]
Tokenizing Prompts (num_proc=192):  96%|█████████▌| 1340272/1393783 [04:16<00:13, 3872.75 examples/s]
Tokenizing Prompts (num_proc=192):  96%|█████████▋| 1341531/1393783 [04:16<00:11, 4362.83 examples/s]
Tokenizing Prompts (num_proc=192):  96%|█████████▋| 1343050/1393783 [04:17<00:16, 3120.70 examples/s]
Tokenizing Prompts (num_proc=192):  96%|█████████▋| 1344309/1393783 [04:17<00:13, 3703.61 examples/s]
Tokenizing Prompts (num_proc=192):  97%|█████████▋| 1345828/1393783 [04:17<00:10, 4467.27 examples/s]
Tokenizing Prompts (num_proc=192):  97%|█████████▋| 1346606/1393783 [04:17<00:11, 4026.60 examples/s]
Tokenizing Prompts (num_proc=192):  97%|█████████▋| 1347866/13937
+1: 83 [04:18<00:10, 4243.55 examples/s]
Tokenizing Prompts (num_proc=192):  97%|█████████▋| 1350126/1393783 [04:18<00:07, 5577.89 examples/s]
Tokenizing Prompts (num_proc=192):  97%|█████████▋| 1351644/1393783 [04:19<00:11, 3530.08 examples/s]
Tokenizing Prompts (num_proc=192):  97%|█████████▋| 1353644/1393783 [04:19<00:09, 4459.43 examples/s]
Tokenizing Prompts (num_proc=192):  97%|█████████▋| 1354903/1393783 [04:19<00:08, 4502.37 examples/s]
Tokenizing Prompts (num_proc=192):  97%|█████████▋| 1355903/1393783 [04:19<00:07, 4895.89 examples/s]
Tokenizing Prompts (num_proc=192):  97%|█████████▋| 1356903/1393783 [04:19<00:07, 5094.73 examples/s]
Tokenizing Prompts (num_proc=192):  97%|█████████▋| 1358162/1393783 [04:20<00:09, 3745.31 examples/s]
Tokenizing Prompts (num_proc=192):  98%|█████████▊| 1360162/1393783 [04:20<00:06, 4944.65 examples/s]
Tokenizing 
+1: Prompts (num_proc=192):  98%|█████████▊| 1361681/1393783 [04:21<00:06, 4587.50 examples/s]
Tokenizing Prompts (num_proc=192):  98%|█████████▊| 1362940/1393783 [04:21<00:07, 3883.98 examples/s]
Tokenizing Prompts (num_proc=192):  98%|█████████▊| 1363459/1393783 [04:21<00:08, 3436.75 examples/s]
Tokenizing Prompts (num_proc=192):  98%|█████████▊| 1363977/1393783 [04:22<00:09, 3101.05 examples/s]
Tokenizing Prompts (num_proc=192):  98%|█████████▊| 1367977/1393783 [04:22<00:05, 5066.90 examples/s]
Tokenizing Prompts (num_proc=192):  98%|█████████▊| 1368495/1393783 [04:22<00:05, 4307.89 examples/s]
Tokenizing Prompts (num_proc=192):  98%|█████████▊| 1369495/1393783 [04:23<00:05, 4432.61 examples/s]
Tokenizing Prompts (num_proc=192):  98%|█████████▊| 1370495/1393783 [04:23<00:05, 4385.59 examples/s]
Tokenizing Prompts (num_proc=192):  98%|██████�
+1: ��██▊| 1371495/1393783 [04:23<00:04, 4610.55 examples/s]
Tokenizing Prompts (num_proc=192):  98%|█████████▊| 1372754/1393783 [04:23<00:04, 4334.05 examples/s]
Tokenizing Prompts (num_proc=192):  99%|█████████▊| 1374272/1393783 [04:23<00:03, 5112.26 examples/s]
Tokenizing Prompts (num_proc=192):  99%|█████████▊| 1375790/1393783 [04:24<00:03, 5966.70 examples/s]
Tokenizing Prompts (num_proc=192):  99%|█████████▉| 1377049/1393783 [04:24<00:03, 4241.67 examples/s]
Tokenizing Prompts (num_proc=192):  99%|█████████▉| 1378308/1393783 [04:24<00:02, 5222.57 examples/s]
Tokenizing Prompts (num_proc=192):  99%|█████████▉| 1380308/1393783 [04:24<00:01, 7089.02 examples/s]
Tokenizing Prompts (num_proc=192):  99%|█████████▉| 1381826/1393783 [04:25<00:01, 6119.38 examples/s]
Tokenizing Prompts (num_proc=192):  99%|█████████▉| 1383085/1393783 [04:25<00:02, 4695.
+1: 30 examples/s]
Tokenizing Prompts (num_proc=192):  99%|█████████▉| 1384344/1393783 [04:26<00:02, 4014.25 examples/s]
Tokenizing Prompts (num_proc=192):  99%|███��█████▉| 1385603/1393783 [04:26<00:01, 4610.28 examples/s]
Tokenizing Prompts (num_proc=192): 100%|█████████▉| 1386862/1393783 [04:26<00:01, 4880.97 examples/s]
Tokenizing Prompts (num_proc=192): 100%|█████████▉| 1387639/1393783 [04:26<00:01, 4261.85 examples/s]
Tokenizing Prompts (num_proc=192): 100%|█████████▉| 1388157/1393783 [04:27<00:02, 2168.25 examples/s]
Tokenizing Prompts (num_proc=192): 100%|█████████▉| 1388675/1393783 [04:28<00:03, 1654.47 examples/s]
Tokenizing Prompts (num_proc=192): 100%|█████████▉| 1389193/1393783 [04:28<00:02, 1842.41 examples/s]
Tokenizing Prompts (num_proc=192): 100%|█████████▉| 1389711/1393783 [04:28<00:01, 2058.51 examples/s]
Tokenizing Prompts (num_proc=192)
+1: : 100%|█████████▉| 1390229/1393783 [04:28<00:01, 1883.82 examples/s]
Tokenizing Prompts (num_proc=192): 100%|█████████▉| 1391488/1393783 [04:29<00:01, 2187.09 examples/s]
Tokenizing Prompts (num_proc=192): 100%|█████████▉| 1392488/1393783 [04:29<00:00, 2538.79 examples/s]
Tokenizing Prompts (num_proc=192): 100%|█████████▉| 1393006/1393783 [04:30<00:00, 1941.55 examples/s]
Tokenizing Prompts (num_proc=192): 100%|█████████▉| 1393524/1393783 [04:32<00:00, 798.58 examples/s] 
Tokenizing Prompts (num_proc=192): 100%|██████████| 1393783/1393783 [04:32<00:00, 788.28 examples/s]
Tokenizing Prompts (num_proc=192): 100%|██████████| 1393783/1393783 [04:33<00:00, 5104.18 examples/s]
+1: 
Dropping Long Sequences (>16384) (num_proc=192):   0%|          | 0/1393783 [00:00<?, ? examples/s]
Dropping Long Sequences (>16384) (num_proc=192):   0%|          | 1000/1393783 [00:03<1:15:21, 308.01 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):   1%|          | 9000/1393783 [00:03<06:18, 3663.31 examples/s] 
Dropping Long Sequences (>16384) (num_proc=192):   1%|          | 17000/1393783 [00:03<02:53, 7956.95 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):   2%|▏         | 28000/1393783 [00:03<01:28, 15457.07 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):   3%|▎         | 39000/1393783 [00:03<00:55, 24352.33 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):   3%|▎         | 48000/1393783 [00:03<00:44, 30345.14 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):   4%|▍         | 56000/1393783 [00:04<00:40, 33304.18 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):   5%|▍         | 63000/1393783 [00:04<00:35, 37685.02 examples
+1: /s]
Dropping Long Sequences (>16384) (num_proc=192):   5%|▌         | 70000/1393783 [00:04<00:37, 35123.55 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):   5%|▌         | 76000/1393783 [00:04<00:34, 38499.55 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):   6%|▌         | 82000/1393783 [00:04<00:36, 36050.94 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):   7%|▋         | 91000/1393783 [00:04<00:28, 45807.76 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):   7%|▋         | 100000/1393783 [00:04<00:24, 52347.60 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):   8%|▊         | 110000/1393783 [00:05<00:20, 61515.77 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):   9%|▊         | 119000/1393783 [00:05<00:18, 67561.43 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):   9%|▉         | 127000/1393783 [00:05<00:19, 66665.42 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  10%|▉         | 135000/13937
+1: 83 [00:05<00:18, 69231.58 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  10%|█         | 143000/1393783 [00:05<00:18, 68328.69 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  11%|█▏        | 158000/1393783 [00:05<00:13, 89344.28 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  12%|█▏        | 172000/1393783 [00:05<00:12, 100685.92 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  13%|█▎        | 186000/1393783 [00:05<00:11, 108263.37 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  14%|█▍        | 198000/1393783 [00:05<00:10, 110210.28 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  15%|█▌        | 216000/1393783 [00:06<00:09, 125808.58 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  16%|█▋        | 229000/1393783 [00:06<00:10, 115097.03 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  17%|█▋        | 241000/1393783 [00:06<00:12, 91900.09 examples/s] 
Dropping Long Sequence
+1: s (>16384) (num_proc=192):  18%|█▊        | 254000/1393783 [00:06<00:14, 81309.70 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  19%|█▉        | 263000/1393783 [00:06<00:14, 75891.65 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  20%|█▉        | 272000/1393783 [00:06<00:18, 60345.67 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  20%|██        | 279000/1393783 [00:07<00:19, 57297.43 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  21%|██        | 286000/1393783 [00:07<00:19, 57839.51 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  21%|██        | 295000/1393783 [00:07<00:17, 64434.57 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  22%|██▏       | 311000/1393783 [00:07<00:13, 80963.23 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  23%|██▎       | 320000/1393783 [00:07<00:14, 72986.29 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  24%|██▎       | 330000/139
+1: 3783 [00:07<00:14, 75242.70 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  25%|██▍       | 342000/1393783 [00:07<00:12, 83033.20 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  26%|██▌       | 358000/1393783 [00:07<00:10, 94675.54 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  27%|██▋       | 373000/1393783 [00:08<00:09, 106771.83 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  28%|██▊       | 390000/1393783 [00:08<00:08, 121905.94 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  29%|██▉       | 403000/1393783 [00:08<00:08, 113675.16 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  30%|██▉       | 415000/1393783 [00:08<00:08, 108820.98 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  31%|███       | 427000/1393783 [00:08<00:11, 85227.38 examples/s] 
Dropping Long Sequences (>16384) (num_proc=192):  31%|███▏      | 437000/1393783 [00:08<00:11, 82652.19 examples/s]
D
+1: ropping Long Sequences (>16384) (num_proc=192):  32%|███▏      | 446000/1393783 [00:08<00:11, 79575.10 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  33%|███▎      | 455000/1393783 [00:09<00:13, 68759.10 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  33%|███▎      | 463000/1393783 [00:09<00:16, 55618.54 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  34%|███▎      | 470000/1393783 [00:09<00:16, 57171.59 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  34%|███▍      | 477000/1393783 [00:09<00:16, 57097.10 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  35%|███▌      | 491000/1393783 [00:09<00:12, 75187.18 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  36%|███▌      | 500000/1393783 [00:09<00:11, 77037.41 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  37%|███▋      | 510000/1393783 [00:09<00:11, 78896.28 examples/s]
Dropping Long Sequences (>16384) 
+1: (num_proc=192):  37%|███▋      | 519000/1393783 [00:09<00:10, 79867.25 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  38%|███▊      | 532000/1393783 [00:10<00:10, 85446.67 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  39%|███▉      | 545000/1393783 [00:10<00:08, 94496.65 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  40%|███▉      | 557000/1393783 [00:10<00:08, 100077.59 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  41%|████      | 568000/1393783 [00:10<00:08, 101379.29 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  42%|████▏     | 587000/1393783 [00:10<00:06, 125445.32 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  43%|████▎     | 600000/1393783 [00:10<00:06, 115282.11 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  44%|████▍     | 612000/1393783 [00:10<00:07, 104360.58 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  45%|
+1: ████▍     | 623000/1393783 [00:11<00:08, 87372.10 examples/s] 
Dropping Long Sequences (>16384) (num_proc=192):  45%|████▌     | 633000/1393783 [00:11<00:09, 78317.99 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  46%|████▌     | 642000/1393783 [00:11<00:11, 66998.57 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  47%|████▋     | 650000/1393783 [00:11<00:11, 65844.89 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  47%|████▋     | 657000/1393783 [00:11<00:13, 56535.17 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  48%|████▊     | 667000/1393783 [00:11<00:11, 64515.88 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  49%|████▊     | 677000/1393783 [00:11<00:10, 71071.21 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  49%|████▉     | 685000/1393783 [00:12<00:10, 66745.48 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  50%|████▉
+1:      | 693000/1393783 [00:12<00:10, 69505.37 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  50%|█████     | 703000/1393783 [00:12<00:09, 72489.92 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  51%|█████▏    | 715000/1393783 [00:12<00:08, 83749.96 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  52%|█████▏    | 726000/1393783 [00:12<00:07, 88686.74 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  53%|█████▎    | 737000/1393783 [00:12<00:07, 92980.40 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  54%|█████▍    | 751000/1393783 [00:12<00:06, 105704.07 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  55%|█████▍    | 762000/1393783 [00:12<00:06, 99562.26 examples/s] 
Dropping Long Sequences (>16384) (num_proc=192):  56%|█████▌    | 778000/1393783 [00:12<00:05, 113264.42 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  57%|█████�
+1: ��    | 790000/1393783 [00:13<00:05, 103555.06 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  57%|█████▋    | 801000/1393783 [00:13<00:05, 104945.89 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  58%|█████▊    | 812000/1393783 [00:13<00:07, 75410.39 examples/s] 
Dropping Long Sequences (>16384) (num_proc=192):  59%|█████▉    | 823000/1393783 [00:13<00:07, 81258.83 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  60%|█████▉    | 833000/1393783 [00:13<00:07, 79615.55 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  60%|██████    | 842000/1393783 [00:13<00:08, 61948.14 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  61%|██████    | 850000/1393783 [00:14<00:08, 61522.13 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  62%|██████▏   | 859000/1393783 [00:14<00:07, 67154.54 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  62%|███�
+1: �██▏   | 868000/1393783 [00:14<00:07, 66619.19 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  63%|██████▎   | 876000/1393783 [00:14<00:07, 66747.09 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  63%|██████▎   | 884000/1393783 [00:14<00:07, 67771.17 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  64%|██████▍   | 898000/1393783 [00:14<00:05, 85049.65 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  65%|██████▌   | 911000/1393783 [00:14<00:05, 95032.42 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  66%|██████▌   | 921000/1393783 [00:14<00:05, 90200.73 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  67%|██████▋   | 932000/1393783 [00:14<00:04, 94727.78 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  68%|██████▊   | 948000/1393783 [00:15<00:04, 104801.28 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):
+1:   69%|██████▉   | 962000/1393783 [00:15<00:03, 111277.57 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  70%|██████▉   | 974000/1393783 [00:15<00:04, 95782.78 examples/s] 
Dropping Long Sequences (>16384) (num_proc=192):  71%|███████   | 987000/1393783 [00:15<00:04, 97175.25 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  72%|███████▏  | 998000/1393783 [00:15<00:04, 83098.07 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  72%|███████▏  | 1007000/1393783 [00:15<00:04, 80162.22 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  73%|███████▎  | 1016000/1393783 [00:15<00:04, 77947.37 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  73%|███████▎  | 1024000/1393783 [00:16<00:04, 77771.92 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  74%|███████▍  | 1032000/1393783 [00:16<00:05, 64961.47 examples/s]
Dropping Long Se
+1: quences (>16384) (num_proc=192):  75%|███████▍  | 1039000/1393783 [00:16<00:05, 61622.26 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  75%|███████▌  | 1048000/1393783 [00:16<00:05, 67548.42 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  76%|███████▌  | 1057000/1393783 [00:16<00:04, 70105.04 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  76%|███████▋  | 1065000/1393783 [00:16<00:04, 69123.15 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  77%|███████▋  | 1073000/1393783 [00:16<00:05, 61376.05 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  78%|███████▊  | 1088000/1393783 [00:16<00:03, 80620.61 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  79%|███████▉  | 1100000/1393783 [00:17<00:03, 89906.55 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  80%|███████▉  | 1112000/1393783 [00:17<00:0
+1: 3, 90223.55 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  81%|████████  | 1129000/1393783 [00:17<00:02, 108153.46 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  82%|████████▏ | 1141000/1393783 [00:17<00:02, 101542.29 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  83%|████████▎ | 1153259/1393783 [00:17<00:02, 101163.42 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  84%|████████▎ | 1164297/1393783 [00:17<00:02, 98999.73 examples/s] 
Dropping Long Sequences (>16384) (num_proc=192):  84%|████████▍ | 1174594/1393783 [00:17<00:02, 92686.96 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  85%|████████▌ | 1186191/1393783 [00:17<00:02, 98558.06 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  86%|████████▌ | 1197710/1393783 [00:18<00:01, 98391.10 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  87%
+1: |████████▋ | 1207784/1393783 [00:18<00:01, 98985.52 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  87%|████████▋ | 1218341/1393783 [00:18<00:01, 96713.93 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  88%|████████▊ | 1228640/1393783 [00:18<00:01, 91407.24 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  89%|████████▉ | 1240457/1393783 [00:18<00:01, 97396.87 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  90%|████████▉ | 1252273/1393783 [00:18<00:01, 102134.09 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  91%|█████████ | 1264611/1393783 [00:18<00:01, 104116.28 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  92%|█████████▏| 1280166/1393783 [00:18<00:00, 116974.21 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  93%|█████████▎| 1294795/1393783 [00:18<00:00, 124889.40 e
+1: xamples/s]
Dropping Long Sequences (>16384) (num_proc=192):  94%|█████████▍| 1308909/1393783 [00:18<00:00, 129451.85 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  95%|█████████▍| 1322019/1393783 [00:19<00:00, 126984.21 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  96%|█████████▌| 1335868/1393783 [00:19<00:00, 128478.69 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  97%|█████████▋| 1349237/1393783 [00:19<00:00, 120699.29 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  98%|█████████▊| 1362865/1393783 [00:19<00:00, 123995.91 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  99%|█████████▊| 1375531/1393783 [00:19<00:00, 106603.46 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  99%|█████████▉| 1386675/1393783 [00:19<00:00, 89054.18 examples/s] 
Dropping Long Sequences (>16384) (num_proc=192)
+1: : 100%|██████████| 1393783/1393783 [00:20<00:00, 67644.96 examples/s]
+1: 
Drop Samples with Zero Trainable Tokens (num_proc=192):   0%|          | 0/1386595 [00:00<?, ? examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   0%|          | 1000/1386595 [00:02<59:15, 389.67 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   1%|          | 9000/1386595 [00:02<05:04, 4520.02 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   1%|▏         | 19000/1386595 [00:02<02:07, 10739.60 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   2%|▏         | 26000/1386595 [00:02<01:26, 15731.92 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   2%|▏         | 33000/1386595 [00:03<01:02, 21615.54 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   3%|▎         | 40000/1386595 [00:03<00:48, 27732.42 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   3%|▎         | 47000/1386595 [00:03<00:42, 31329.27 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  
+1:  4%|▍         | 54000/1386595 [00:03<00:35, 37695.59 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   4%|▍         | 60000/1386595 [00:03<00:34, 38083.77 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   5%|▌         | 73000/1386595 [00:03<00:23, 56551.25 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   6%|▌         | 84000/1386595 [00:03<00:19, 66969.57 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   7%|▋         | 93000/1386595 [00:03<00:19, 65640.87 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   7%|▋         | 102000/1386595 [00:04<00:18, 69556.79 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   8%|▊         | 110000/1386595 [00:04<00:20, 62237.99 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   8%|▊         | 117000/1386595 [00:04<00:26, 48613.89 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   9%|▉         | 124000/1386
+1: 595 [00:04<00:24, 51062.46 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  10%|▉         | 132000/1386595 [00:04<00:23, 53869.56 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  10%|█         | 139000/1386595 [00:04<00:22, 56414.36 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  11%|█         | 146000/1386595 [00:04<00:22, 54577.52 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  12%|█▏        | 160000/1386595 [00:05<00:16, 72310.28 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  12%|█▏        | 168000/1386595 [00:05<00:20, 60173.66 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  13%|█▎        | 175000/1386595 [00:05<00:21, 56686.69 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  13%|█▎        | 185000/1386595 [00:05<00:18, 64911.75 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  14%|█▍        | 192000/1386595 [00:05<00:
+1: 19, 61498.70 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  14%|█▍        | 200000/1386595 [00:05<00:18, 64212.91 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  15%|█▌        | 208000/1386595 [00:05<00:18, 65107.82 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  16%|█▌        | 217000/1386595 [00:05<00:16, 71214.61 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  16%|█▌        | 225000/1386595 [00:06<00:15, 73056.47 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  17%|█▋        | 234000/1386595 [00:06<00:15, 73830.72 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  17%|█▋        | 242000/1386595 [00:06<00:15, 74011.49 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  18%|█▊        | 250000/1386595 [00:06<00:15, 74559.87 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  19%|█▊        | 258000/1386595 [00:06<00:20, 5555
+1: 2.07 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  19%|█▉        | 269000/1386595 [00:06<00:16, 67540.15 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  20%|██        | 281000/1386595 [00:06<00:14, 75928.23 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  21%|██        | 292000/1386595 [00:06<00:13, 81482.00 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  22%|██▏       | 306000/1386595 [00:07<00:11, 94365.65 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  23%|██▎       | 316000/1386595 [00:07<00:11, 89681.51 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  24%|██▍       | 330000/1386595 [00:07<00:10, 98256.93 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  25%|██▍       | 341000/1386595 [00:07<00:12, 83160.86 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  25%|██▌       | 350000/1386595 [00:07<00:13, 78
+1: 321.61 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  26%|██▌       | 362000/1386595 [00:07<00:11, 87292.08 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  27%|██▋       | 375000/1386595 [00:07<00:11, 91727.25 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  28%|██▊       | 385000/1386595 [00:08<00:12, 81572.79 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  29%|██▊       | 396000/1386595 [00:08<00:11, 88130.75 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  29%|██▉       | 406000/1386595 [00:08<00:10, 89328.93 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  30%|███       | 416000/1386595 [00:08<00:10, 88361.98 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  31%|███       | 426000/1386595 [00:08<00:11, 80183.83 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  32%|███▏      | 437000/1386595 [00:08
+1: <00:11, 86262.56 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  32%|███▏      | 446000/1386595 [00:08<00:12, 73424.52 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  33%|███▎      | 454000/1386595 [00:08<00:14, 63696.49 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  33%|███▎      | 464000/1386595 [00:09<00:13, 69981.15 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  34%|███▍      | 472000/1386595 [00:09<00:13, 70263.40 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  35%|███▌      | 489000/1386595 [00:09<00:10, 89399.96 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  36%|███▌      | 501000/1386595 [00:09<00:09, 95474.66 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  37%|███▋      | 514000/1386595 [00:09<00:08, 103895.22 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  38%|███▊    
+1:   | 525000/1386595 [00:09<00:08, 100646.01 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  39%|███▊      | 536000/1386595 [00:09<00:11, 74915.43 examples/s] 
Drop Samples with Zero Trainable Tokens (num_proc=192):  39%|███▉      | 546000/1386595 [00:10<00:10, 79013.96 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  40%|████      | 560000/1386595 [00:10<00:09, 90275.58 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  41%|████      | 570000/1386595 [00:10<00:09, 86337.42 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  42%|████▏     | 580000/1386595 [00:10<00:09, 84282.35 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  43%|████▎     | 593000/1386595 [00:10<00:08, 93633.94 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  43%|████▎     | 603000/1386595 [00:10<00:09, 81445.67 examples/s]
Drop Samples with Zero Trainable Tokens (num_
+1: proc=192):  44%|████▍     | 617000/1386595 [00:10<00:08, 94505.85 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  45%|████▌     | 628000/1386595 [00:10<00:09, 77002.65 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  46%|████▌     | 637000/1386595 [00:11<00:11, 67972.61 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  47%|████▋     | 645000/1386595 [00:11<00:11, 65470.82 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  47%|████▋     | 658000/1386595 [00:11<00:09, 78958.27 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  48%|████▊     | 667000/1386595 [00:11<00:09, 75373.37 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  49%|████▉     | 676000/1386595 [00:11<00:09, 76188.20 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  50%|████▉     | 691000/1386595 [00:11<00:07, 93663.12 examples/s]
Drop 
+1: Samples with Zero Trainable Tokens (num_proc=192):  51%|█████     | 707000/1386595 [00:11<00:06, 106544.44 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  52%|█████▏    | 719000/1386595 [00:12<00:07, 91117.46 examples/s] 
Drop Samples with Zero Trainable Tokens (num_proc=192):  53%|█████▎    | 729000/1386595 [00:12<00:07, 82426.14 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  53%|█████▎    | 738000/1386595 [00:12<00:07, 82412.51 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  54%|█████▍    | 750000/1386595 [00:12<00:06, 90970.53 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  55%|█████▍    | 760000/1386595 [00:12<00:07, 83340.64 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  55%|█████▌    | 769000/1386595 [00:12<00:08, 75989.49 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  56%|█████▋    |
+1:  783000/1386595 [00:12<00:06, 89851.06 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  57%|█████▋    | 793000/1386595 [00:12<00:06, 88065.72 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  58%|█████▊    | 807000/1386595 [00:13<00:06, 92901.40 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  59%|█████▉    | 817000/1386595 [00:13<00:06, 91748.48 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  60%|█████▉    | 827000/1386595 [00:13<00:07, 74514.10 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  60%|██████    | 835222/1386595 [00:13<00:08, 65949.08 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  61%|██████    | 843222/1386595 [00:13<00:07, 68505.45 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  62%|██████▏   | 853222/1386595 [00:13<00:07, 74957.83 examples/s]
Drop Samples with Zero Tra
+1: inable Tokens (num_proc=192):  62%|██████▏   | 862222/1386595 [00:13<00:07, 72306.51 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  63%|██████▎   | 876222/1386595 [00:13<00:05, 87705.55 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  64%|██████▍   | 889222/1386595 [00:14<00:05, 97215.54 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  65%|██████▍   | 900222/1386595 [00:14<00:04, 99567.05 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  66%|██████▌   | 911444/1386595 [00:14<00:04, 101493.68 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  67%|██████▋   | 922666/1386595 [00:14<00:05, 82212.97 examples/s] 
Drop Samples with Zero Trainable Tokens (num_proc=192):  67%|██████▋   | 934110/1386595 [00:14<00:05, 89119.23 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  68%|██████▊   | 94
+1: 4110/1386595 [00:14<00:04, 90854.62 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  69%|██████▉   | 954554/1386595 [00:14<00:05, 75417.96 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  70%|██████▉   | 963776/1386595 [00:15<00:05, 76977.73 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  70%|███████   | 976776/1386595 [00:15<00:04, 89146.69 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  71%|███████   | 986998/1386595 [00:15<00:04, 84284.27 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  72%|███████▏  | 999220/1386595 [00:15<00:04, 93530.38 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  73%|███████▎  | 1012664/1386595 [00:15<00:03, 95916.39 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  74%|███████▍  | 1022886/1386595 [00:15<00:04, 88274.61 examples/s]
Drop Samp
+1: les with Zero Trainable Tokens (num_proc=192):  74%|███████▍  | 1032108/1386595 [00:15<00:04, 87295.16 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  75%|███████▌  | 1041884/1386595 [00:15<00:04, 72270.90 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  76%|███████▌  | 1050106/1386595 [00:16<00:04, 72983.56 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  77%|███████▋  | 1061550/1386595 [00:16<00:03, 82184.55 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  77%|███████▋  | 1073550/1386595 [00:16<00:03, 90210.95 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  78%|███████▊  | 1086994/1386595 [00:16<00:02, 101739.17 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  79%|███████▉  | 1101438/1386595 [00:16<00:02, 111861.96 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=1
+1: 92):  80%|████████  | 1113104/1386595 [00:16<00:02, 109638.93 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  81%|████████  | 1124658/1386595 [00:16<00:02, 98030.43 examples/s] 
Drop Samples with Zero Trainable Tokens (num_proc=192):  82%|████████▏ | 1137546/1386595 [00:16<00:02, 104068.81 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  83%|████████▎ | 1148656/1386595 [00:16<00:02, 101112.83 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  84%|████████▎ | 1159210/1386595 [00:17<00:02, 101402.85 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  84%|████████▍ | 1170432/1386595 [00:17<00:02, 101628.24 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  85%|████████▌ | 1181320/1386595 [00:17<00:02, 98522.41 examples/s] 
Drop Samples with Zero Trainable Tokens (num_proc=192):  86%|█████�
+1: �██▌ | 1192096/1386595 [00:17<00:01, 99188.63 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  87%|████████▋ | 1205984/1386595 [00:17<00:01, 106099.60 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  88%|████████▊ | 1217316/1386595 [00:17<00:01, 99966.99 examples/s] 
Drop Samples with Zero Trainable Tokens (num_proc=192):  89%|████████▉ | 1233204/1386595 [00:17<00:01, 114994.10 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  90%|████████▉ | 1245980/1386595 [00:17<00:01, 118416.33 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  91%|█████████ | 1258200/1386595 [00:18<00:01, 97611.36 examples/s] 
Drop Samples with Zero Trainable Tokens (num_proc=192):  92%|█████████▏| 1271087/1386595 [00:18<00:01, 104967.53 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  93%|█████████▎| 1282974
+1: /1386595 [00:18<00:01, 98348.72 examples/s] 
Drop Samples with Zero Trainable Tokens (num_proc=192):  93%|█████████▎| 1293528/1386595 [00:18<00:00, 94151.49 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  94%|█████████▍| 1303416/1386595 [00:18<00:00, 91065.69 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  95%|█████████▍| 1313524/1386595 [00:18<00:00, 91339.39 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  95%|█████████▌| 1322856/1386595 [00:18<00:00, 86551.73 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  96%|█████████▋| 1336851/1386595 [00:18<00:00, 99806.51 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  97%|█████████▋| 1347183/1386595 [00:18<00:00, 96603.38 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  98%|█████████▊| 1357843/1386595 [00:19
+1: <00:00, 94367.25 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  99%|█████████▊| 1368389/1386595 [00:19<00:00, 74124.77 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  99%|█████████▉| 1377496/1386595 [00:19<00:00, 59428.01 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192): 100%|█████████▉| 1384488/1386595 [00:19<00:00, 48068.05 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192): 100%|██████████| 1386595/1386595 [00:20<00:00, 66850.22 examples/s]
+1: 
Add position_id column (Sample Packing) (num_proc=192):   0%|          | 0/1386595 [00:00<?, ? examples/s]
Add position_id column (Sample Packing) (num_proc=192):   0%|          | 1000/1386595 [00:03<1:24:48, 272.30 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   1%|          | 15000/1386595 [00:03<04:09, 5496.26 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   2%|▏         | 23000/1386595 [00:03<02:29, 9149.44 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   2%|▏         | 33000/1386595 [00:04<01:31, 14840.76 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   3%|▎         | 41000/1386595 [00:04<01:07, 19891.15 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   4%|▎         | 49000/1386595 [00:04<00:51, 25785.09 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   4%|▍         | 57000/1386595 [00:04<00:40, 32770.62 examples/s]
Add position_id column (Sample Packing) (num_proc=192):
+1:    5%|▍         | 65000/1386595 [00:04<00:33, 39110.08 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   5%|▌         | 73000/1386595 [00:04<00:37, 34571.36 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   6%|▌         | 79000/1386595 [00:05<00:43, 29857.38 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   6%|▌         | 84000/1386595 [00:05<00:43, 29933.72 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   6%|▋         | 89000/1386595 [00:05<00:42, 30780.66 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   7%|▋         | 93000/1386595 [00:05<00:51, 25231.16 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   7%|▋         | 98000/1386595 [00:05<00:48, 26632.69 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   8%|▊         | 105000/1386595 [00:05<00:38, 32899.80 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   8%|▊         | 113000/1386
+1: 595 [00:06<00:31, 40508.84 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   9%|▊         | 119000/1386595 [00:06<00:29, 43696.10 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  10%|▉         | 132000/1386595 [00:06<00:25, 49364.38 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  10%|▉         | 138000/1386595 [00:06<00:25, 48152.24 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  10%|█         | 145000/1386595 [00:06<00:25, 48742.84 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  11%|█         | 150000/1386595 [00:06<00:25, 48138.09 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  11%|█         | 155000/1386595 [00:06<00:26, 45911.78 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  12%|█▏        | 160000/1386595 [00:07<00:35, 34634.95 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  12%|█▏        | 167000/1386595 [00:07<00:30, 40
+1: 241.23 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  12%|█▏        | 172000/1386595 [00:07<00:30, 39452.30 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  13%|█▎        | 185000/1386595 [00:07<00:20, 59505.74 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  14%|█▍        | 193000/1386595 [00:07<00:19, 62579.19 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  14%|█▍        | 201000/1386595 [00:07<00:17, 66902.25 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  15%|█▌        | 209000/1386595 [00:07<00:16, 70162.36 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  16%|█▌        | 225000/1386595 [00:07<00:13, 85905.40 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  17%|█▋        | 235000/1386595 [00:08<00:13, 87966.98 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  18%|█▊        | 249000/1386595 [00:08<00:11, 98042.92 e
+1: xamples/s]
Add position_id column (Sample Packing) (num_proc=192):  19%|█▊        | 259000/1386595 [00:08<00:11, 97876.56 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  19%|█▉        | 269000/1386595 [00:08<00:19, 58811.21 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  20%|█▉        | 277000/1386595 [00:08<00:24, 44803.44 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  20%|██        | 284000/1386595 [00:09<00:28, 38874.78 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  21%|██        | 290000/1386595 [00:09<00:31, 35062.60 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  21%|██▏       | 296000/1386595 [00:09<00:30, 35313.01 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  22%|██▏       | 301000/1386595 [00:09<00:29, 37057.51 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  22%|██▏       | 307000/1386595 [00:09<00:26, 41124.75 exa
+1: mples/s]
Add position_id column (Sample Packing) (num_proc=192):  23%|██▎       | 320000/1386595 [00:09<00:18, 56864.95 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  24%|██▎       | 327000/1386595 [00:10<00:20, 52196.45 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  24%|██▍       | 333000/1386595 [00:10<00:25, 41717.25 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  25%|██▍       | 342000/1386595 [00:10<00:20, 49779.24 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  25%|██▌       | 348000/1386595 [00:10<00:26, 38998.93 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  26%|██▌       | 354000/1386595 [00:10<00:24, 41988.46 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  26%|██▌       | 360000/1386595 [00:10<00:23, 42873.07 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  26%|██▋       | 366000/1386595 [00:11<00:22, 4488
+1: 9.78 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  27%|██▋       | 375000/1386595 [00:11<00:18, 54092.47 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  28%|██▊       | 384000/1386595 [00:11<00:17, 58973.71 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  28%|██▊       | 393000/1386595 [00:11<00:15, 64665.94 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  29%|██▉       | 400000/1386595 [00:11<00:15, 63200.85 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  30%|██▉       | 410000/1386595 [00:11<00:13, 71459.23 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  30%|███       | 421000/1386595 [00:11<00:12, 78290.52 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  32%|███▏      | 437000/1386595 [00:11<00:09, 95080.87 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  32%|███▏      | 448000/1386595 [00:11
+1: <00:09, 98448.97 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  33%|███▎      | 458000/1386595 [00:12<00:13, 70453.17 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  34%|███▎      | 467000/1386595 [00:12<00:19, 47904.34 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  34%|███▍      | 474000/1386595 [00:12<00:22, 40733.06 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  35%|███▍      | 480000/1386595 [00:13<00:25, 36136.03 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  35%|███▍      | 485000/1386595 [00:13<00:25, 34891.11 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  35%|███▌      | 491000/1386595 [00:13<00:23, 37491.57 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  36%|███▌      | 496000/1386595 [00:13<00:23, 38331.68 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  36%|███▋     
+1:  | 504000/1386595 [00:13<00:20, 42348.31 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  37%|███▋      | 516000/1386595 [00:13<00:17, 51101.71 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  38%|███▊      | 522000/1386595 [00:14<00:19, 43918.13 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  38%|███▊      | 527000/1386595 [00:14<00:19, 43410.39 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  39%|███▊      | 534000/1386595 [00:14<00:19, 44146.59 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  39%|███▉      | 541000/1386595 [00:14<00:17, 48999.07 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  39%|███▉      | 547000/1386595 [00:14<00:17, 47327.44 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  40%|███▉      | 552000/1386595 [00:14<00:18, 46106.44 examples/s]
Add position_id column (Sample Packing) (num_proc=192)
+1: :  40%|████      | 559000/1386595 [00:14<00:18, 45188.67 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  41%|████      | 565000/1386595 [00:14<00:17, 46687.41 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  41%|████▏     | 575000/1386595 [00:15<00:14, 54928.91 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  43%|████▎     | 590000/1386595 [00:15<00:10, 76468.47 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  43%|████▎     | 599000/1386595 [00:15<00:11, 66258.69 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  44%|████▍     | 612000/1386595 [00:15<00:09, 78366.65 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  45%|████▍     | 621000/1386595 [00:15<00:10, 75541.50 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  46%|████▌     | 632000/1386595 [00:15<00:08, 83867.15 examples/s]
Add position_id co
+1: lumn (Sample Packing) (num_proc=192):  46%|████▌     | 641000/1386595 [00:15<00:08, 83444.78 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  47%|████▋     | 650000/1386595 [00:15<00:10, 72167.28 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  47%|████▋     | 658000/1386595 [00:16<00:11, 64258.78 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  48%|████▊     | 665000/1386595 [00:16<00:19, 36376.60 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  49%|████▊     | 673000/1386595 [00:16<00:17, 40510.39 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  49%|████▉     | 679000/1386595 [00:16<00:17, 39756.92 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  49%|████▉     | 685000/1386595 [00:17<00:17, 40653.67 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  50%|████▉     | 690000/1386595 [00:17<00:16,
+1:  42430.09 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  50%|█████     | 695000/1386595 [00:17<00:16, 41252.62 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  51%|█████     | 704000/1386595 [00:17<00:13, 50432.55 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  51%|█████▏    | 711000/1386595 [00:17<00:13, 51727.40 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  52%|█████▏    | 717000/1386595 [00:17<00:15, 42880.91 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  52%|█████▏    | 722000/1386595 [00:17<00:15, 43287.70 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  53%|█████▎    | 729000/1386595 [00:17<00:13, 47500.85 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  53%|█████▎    | 735000/1386595 [00:18<00:14, 43930.94 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  53%|
+1: █████▎    | 740000/1386595 [00:18<00:15, 40880.73 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  54%|█████▍    | 746000/1386595 [00:18<00:15, 42382.54 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  54%|█████▍    | 752000/1386595 [00:18<00:13, 45884.13 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  55%|█████▍    | 759000/1386595 [00:18<00:12, 50899.00 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  55%|█████▌    | 765000/1386595 [00:18<00:13, 47503.42 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  56%|█████▌    | 775000/1386595 [00:18<00:10, 57963.34 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  57%|█████▋    | 785000/1386595 [00:18<00:08, 68268.73 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  57%|█████▋    | 797000/1386595 [00:19<00:07, 81379.39 examples/s]
Add p
+1: osition_id column (Sample Packing) (num_proc=192):  58%|█████▊    | 806000/1386595 [00:19<00:07, 76527.47 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  59%|█████▉    | 817000/1386595 [00:19<00:07, 78999.80 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  60%|█████▉    | 826000/1386595 [00:19<00:07, 79289.47 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  60%|██████    | 835000/1386595 [00:19<00:07, 70735.73 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  61%|██████    | 843000/1386595 [00:19<00:07, 72776.03 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  61%|██████▏   | 851000/1386595 [00:19<00:09, 57949.51 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  62%|██████▏   | 858000/1386595 [00:20<00:13, 39698.22 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  62%|██████�
+1: ��   | 866000/1386595 [00:20<00:12, 43078.98 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  63%|██████▎   | 872000/1386595 [00:20<00:11, 46043.51 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  63%|██████▎   | 878000/1386595 [00:20<00:12, 41631.55 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  64%|██████▎   | 883000/1386595 [00:20<00:12, 39597.46 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  64%|██████▍   | 890000/1386595 [00:20<00:10, 45681.21 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  65%|██████▍   | 897000/1386595 [00:21<00:10, 45822.97 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  65%|██████▌   | 903000/1386595 [00:21<00:10, 44562.78 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  66%|██████▌   | 911000/1386595 [00:21<00:09, 50698.28 examples/s]
Add posi
+1: tion_id column (Sample Packing) (num_proc=192):  66%|██████▌   | 917000/1386595 [00:21<00:12, 36578.74 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  66%|██████▋   | 922000/1386595 [00:21<00:12, 38088.72 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  67%|██████▋   | 928000/1386595 [00:21<00:10, 41958.40 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  67%|██████▋   | 933000/1386595 [00:21<00:10, 42194.37 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  68%|██████▊   | 938000/1386595 [00:22<00:10, 41080.28 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  68%|██████▊   | 944000/1386595 [00:22<00:10, 42710.74 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  69%|██████▊   | 952000/1386595 [00:22<00:08, 51157.23 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  69%|████
+1: ██▉   | 958000/1386595 [00:22<00:08, 49648.15 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  70%|███████   | 971000/1386595 [00:22<00:06, 60683.16 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  71%|███████   | 981000/1386595 [00:22<00:05, 68650.65 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  71%|███████▏  | 990000/1386595 [00:22<00:05, 73550.91 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  72%|███████▏  | 1000000/1386595 [00:22<00:04, 80277.29 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  73%|███████▎  | 1009000/1386595 [00:23<00:05, 66348.74 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  74%|███████▎  | 1022000/1386595 [00:23<00:04, 73257.60 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  74%|███████▍  | 1030000/1386595 [00:23<00:04, 74505.52
+1:  examples/s]
Add position_id column (Sample Packing) (num_proc=192):  75%|███████▍  | 1038000/1386595 [00:23<00:05, 62944.40 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  75%|███████▌  | 1045000/1386595 [00:23<00:06, 51653.24 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  76%|███████▌  | 1051000/1386595 [00:23<00:06, 51670.45 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  76%|███████▌  | 1057000/1386595 [00:23<00:06, 51855.28 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  77%|███████▋  | 1063000/1386595 [00:24<00:06, 47631.42 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  77%|███████▋  | 1068000/1386595 [00:24<00:08, 39490.18 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  77%|███████▋  | 1073000/1386595 [00:24<00:08, 36014.18 examples/s]
Add position_id column (Sample 
+1: Packing) (num_proc=192):  78%|███████▊  | 1082222/1386595 [00:24<00:06, 47037.92 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  78%|███████▊  | 1088222/1386595 [00:24<00:06, 45474.68 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  79%|███████▉  | 1095222/1386595 [00:24<00:06, 48202.85 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  79%|███████▉  | 1101222/1386595 [00:24<00:05, 49370.77 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  80%|███████▉  | 1107222/1386595 [00:25<00:06, 45527.41 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  80%|████████  | 1112222/1386595 [00:25<00:07, 38349.80 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  81%|████████  | 1117222/1386595 [00:25<00:06, 40403.53 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  81%|████�
+1: �███  | 1124222/1386595 [00:25<00:06, 43710.25 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  82%|████████▏ | 1131222/1386595 [00:25<00:05, 43396.30 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  82%|████████▏ | 1136444/1386595 [00:25<00:05, 43601.15 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  82%|████████▏ | 1142110/1386595 [00:25<00:05, 44326.94 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  83%|████████▎ | 1146776/1386595 [00:26<00:05, 41485.70 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  83%|████████▎ | 1156776/1386595 [00:26<00:04, 55700.07 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  84%|████████▍ | 1167442/1386595 [00:26<00:03, 68615.07 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  85%|████████▍ | 1175664/1386595 
+1: [00:26<00:02, 71900.84 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  85%|████████▌ | 1183886/1386595 [00:26<00:02, 73991.28 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  86%|████████▋ | 1196108/1386595 [00:26<00:02, 87305.98 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  87%|████████▋ | 1205218/1386595 [00:26<00:02, 73357.41 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  88%|████████▊ | 1213662/1386595 [00:26<00:02, 71262.24 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  88%|████████▊ | 1223772/1386595 [00:27<00:02, 78796.79 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  89%|████████▉ | 1232992/1386595 [00:27<00:02, 76612.82 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  90%|████████▉ | 1241324/1386595 [00:27<00:01, 75473.96 example
+1: s/s]
Add position_id column (Sample Packing) (num_proc=192):  90%|█████████ | 1249766/1386595 [00:27<00:02, 63423.79 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  91%|█████████ | 1260098/1386595 [00:27<00:01, 72539.67 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  91%|█████████▏| 1268430/1386595 [00:27<00:01, 70185.17 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  92%|█████████▏| 1276650/1386595 [00:27<00:01, 60978.65 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  93%|█████████▎| 1284538/1386595 [00:27<00:01, 64675.31 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  93%|█████████▎| 1292092/1386595 [00:28<00:01, 63154.98 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  94%|█████████▍| 1301868/1386595 [00:28<00:01, 69777.14 examples/s]
Add position_id
+1:  column (Sample Packing) (num_proc=192):  94%|█████████▍| 1310088/1386595 [00:28<00:01, 72906.34 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  95%|█████████▌| 1317864/1386595 [00:28<00:00, 73878.43 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  96%|█████████▌| 1326418/1386595 [00:28<00:00, 64897.74 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  96%|█████████▌| 1333859/1386595 [00:28<00:00, 65467.28 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  97%|█████████▋| 1343857/1386595 [00:28<00:00, 73533.75 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  98%|█████████▊| 1355189/1386595 [00:28<00:00, 83762.13 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  98%|█████████▊| 1363964/1386595 [00:29<00:00, 70297.00 examples/s]
Add position_id column (Sample 
+1: Packing) (num_proc=192):  99%|█████████▉| 1371621/1386595 [00:29<00:00, 64052.68 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  99%|█████████▉| 1378613/1386595 [00:29<00:00, 62565.76 examples/s]
Add position_id column (Sample Packing) (num_proc=192): 100%|█████████▉| 1385269/1386595 [00:29<00:00, 40680.39 examples/s]
Add position_id column (Sample Packing) (num_proc=192): 100%|██████████| 1386595/1386595 [00:30<00:00, 45600.22 examples/s]
+1: 
Saving the dataset (0/192 shards):   0%|          | 0/1386595 [00:00<?, ? examples/s]
Saving the dataset (0/192 shards):   0%|          | 1000/1386595 [00:02<58:37, 393.90 examples/s]
Saving the dataset (0/192 shards):   3%|▎         | 36000/1386595 [00:02<01:11, 18980.28 examples/s]
Saving the dataset (0/192 shards):  10%|▉         | 132000/1386595 [00:02<00:14, 84860.15 examples/s]
Saving the dataset (0/192 shards):  16%|█▌        | 219000/1386595 [00:02<00:07, 154321.08 examples/s]
Saving the dataset (0/192 shards):  22%|██▏       | 302000/1386595 [00:02<00:04, 228982.74 examples/s]
Saving the dataset (0/192 shards):  27%|██▋       | 369000/1386595 [00:03<00:03, 273897.05 examples/s]
Saving the dataset (0/192 shards):  38%|███▊      | 527000/1386595 [00:03<00:01, 479512.12 examples/s]
Saving the dataset (1/192 shards):  45%|████▍     | 620222/1386595 [00:03<00:01, 479512.12 examples/s]
Saving the dataset (2/192 shards):  51%|█████     | 701444/1386595 [00:03
+1: <00:01, 479512.12 examples/s]
Saving the dataset (2/192 shards):  52%|█████▏    | 714444/1386595 [00:03<00:00, 735587.15 examples/s]
Saving the dataset (3/192 shards):  52%|█████▏    | 716666/1386595 [00:03<00:00, 735587.15 examples/s]
Saving the dataset (4/192 shards):  52%|█████▏    | 720888/1386595 [00:03<00:00, 735587.15 examples/s]
Saving the dataset (5/192 shards):  53%|█████▎    | 730110/1386595 [00:03<00:00, 735587.15 examples/s]
Saving the dataset (6/192 shards):  54%|█████▎    | 744332/1386595 [00:03<00:00, 735587.15 examples/s]
Saving the dataset (7/192 shards):  54%|█████▍    | 752554/1386595 [00:03<00:00, 735587.15 examples/s]
Saving the dataset (8/192 shards):  56%|█████▌    | 777776/1386595 [00:03<00:00, 735587.15 examples/s]
Saving the dataset (9/192 shards):  57%|█████▋    | 790998/1386595 [00:03<00:00, 735587.15 examples/s]
Saving the dataset (10/192 shards):  58%|█████▊    | 810220/13
+1: 86595 [00:03<00:00, 735587.15 examples/s]
Saving the dataset (11/192 shards):  59%|█████▊    | 812442/1386595 [00:03<00:00, 735587.15 examples/s]
Saving the dataset (12/192 shards):  60%|██████    | 835664/1386595 [00:03<00:00, 735587.15 examples/s]
Saving the dataset (12/192 shards):  60%|██████    | 836664/1386595 [00:03<00:00, 678986.72 examples/s]
Saving the dataset (13/192 shards):  62%|██████▏   | 858886/1386595 [00:03<00:00, 678986.72 examples/s]
Saving the dataset (14/192 shards):  62%|██████▏   | 863108/1386595 [00:03<00:00, 678986.72 examples/s]
Saving the dataset (15/192 shards):  63%|██████▎   | 868330/1386595 [00:03<00:00, 678986.72 examples/s]
Saving the dataset (16/192 shards):  64%|██████▎   | 880552/1386595 [00:03<00:00, 678986.72 examples/s]
Saving the dataset (17/192 shards):  65%|██████▍   | 894996/1386595 [00:03<00:00, 678986.72 examples/s]
Saving the dataset (18/192 shards):  65%|█
+1: █████▍   | 894996/1386595 [00:03<00:00, 678986.72 examples/s]
Saving the dataset (19/192 shards):  66%|██████▌   | 910218/1386595 [00:03<00:00, 678986.72 examples/s]
Saving the dataset (20/192 shards):  66%|██████▋   | 919440/1386595 [00:03<00:00, 678986.72 examples/s]
Saving the dataset (21/192 shards):  67%|██████▋   | 925662/1386595 [00:03<00:00, 678986.72 examples/s]
Saving the dataset (22/192 shards):  67%|██████▋   | 928884/1386595 [00:03<00:00, 678986.72 examples/s]
Saving the dataset (23/192 shards):  68%|██████▊   | 937106/1386595 [00:03<00:00, 678986.72 examples/s]
Saving the dataset (23/192 shards):  68%|██████▊   | 939106/1386595 [00:03<00:00, 633711.74 examples/s]
Saving the dataset (24/192 shards):  69%|██████▉   | 954328/1386595 [00:03<00:00, 633711.74 examples/s]
Saving the dataset (25/192 shards):  70%|██████▉   | 967550/1386595 [00:03<00:00, 633711.74 examples/s]
Saving
+1:  the dataset (26/192 shards):  70%|███████   | 976772/1386595 [00:03<00:00, 633711.74 examples/s]
Saving the dataset (27/192 shards):  71%|███████   | 980216/1386595 [00:03<00:00, 633711.74 examples/s]
Saving the dataset (28/192 shards):  71%|███████   | 980216/1386595 [00:03<00:00, 633711.74 examples/s]
Saving the dataset (29/192 shards):  71%|███████   | 981438/1386595 [00:03<00:00, 633711.74 examples/s]
Saving the dataset (30/192 shards):  71%|███████   | 985660/1386595 [00:03<00:00, 633711.74 examples/s]
Saving the dataset (31/192 shards):  71%|███████   | 986104/1386595 [00:03<00:00, 633711.74 examples/s]
Saving the dataset (32/192 shards):  71%|███████   | 987326/1386595 [00:03<00:00, 633711.74 examples/s]
Saving the dataset (33/192 shards):  71%|███████▏  | 988326/1386595 [00:03<00:00, 633711.74 examples/s]
Saving the dataset (34/192 shards):  72%|███████▏  | 991548/1386595 
+1: [00:03<00:00, 633711.74 examples/s]
Saving the dataset (35/192 shards):  72%|███████▏  | 999770/1386595 [00:03<00:00, 633711.74 examples/s]
Saving the dataset (36/192 shards):  73%|███████▎  | 1006992/1386595 [00:03<00:00, 633711.74 examples/s]
Saving the dataset (37/192 shards):  73%|███████▎  | 1012214/1386595 [00:03<00:00, 633711.74 examples/s]
Saving the dataset (38/192 shards):  74%|███████▎  | 1020436/1386595 [00:03<00:00, 633711.74 examples/s]
Saving the dataset (39/192 shards):  74%|███████▎  | 1021658/1386595 [00:03<00:00, 633711.74 examples/s]
Saving the dataset (39/192 shards):  74%|███████▍  | 1026658/1386595 [00:03<00:00, 620689.82 examples/s]
Saving the dataset (40/192 shards):  75%|███████▍  | 1036880/1386595 [00:03<00:00, 620689.82 examples/s]
Saving the dataset (41/192 shards):  75%|███████▍  | 1038102/1386595 [00:03<00:00, 620689.82 examples/s]
Saving the dataset (4
+1: 2/192 shards):  75%|███████▌  | 1044546/1386595 [00:03<00:00, 620689.82 examples/s]
Saving the dataset (43/192 shards):  75%|███████▌  | 1044546/1386595 [00:03<00:00, 620689.82 examples/s]
Saving the dataset (44/192 shards):  76%|███████▌  | 1055768/1386595 [00:03<00:00, 620689.82 examples/s]
Saving the dataset (45/192 shards):  76%|███████▋  | 1057434/1386595 [00:03<00:00, 620689.82 examples/s]
Saving the dataset (46/192 shards):  76%|███████▋  | 1057434/1386595 [00:03<00:00, 620689.82 examples/s]
Saving the dataset (47/192 shards):  76%|███████▋  | 1057434/1386595 [00:03<00:00, 620689.82 examples/s]
Saving the dataset (48/192 shards):  76%|███████▋  | 1059656/1386595 [00:03<00:00, 620689.82 examples/s]
Saving the dataset (49/192 shards):  77%|███████▋  | 1070878/1386595 [00:03<00:00, 620689.82 examples/s]
Saving the dataset (50/192 shards):  78%|███████▊  | 1075100/
+1: 1386595 [00:03<00:00, 620689.82 examples/s]
Saving the dataset (51/192 shards):  78%|███████▊  | 1081322/1386595 [00:03<00:00, 620689.82 examples/s]
Saving the dataset (52/192 shards):  78%|███████▊  | 1083544/1386595 [00:03<00:00, 620689.82 examples/s]
Saving the dataset (53/192 shards):  78%|███████▊  | 1085766/1386595 [00:03<00:00, 620689.82 examples/s]
Saving the dataset (54/192 shards):  78%|███████▊  | 1086210/1386595 [00:03<00:00, 620689.82 examples/s]
Saving the dataset (55/192 shards):  78%|███████▊  | 1086432/1386595 [00:03<00:00, 620689.82 examples/s]
Saving the dataset (56/192 shards):  78%|███████▊  | 1086432/1386595 [00:03<00:00, 620689.82 examples/s]
Saving the dataset (57/192 shards):  79%|███████▊  | 1090876/1386595 [00:03<00:00, 620689.82 examples/s]
Saving the dataset (58/192 shards):  79%|███████▊  | 1091876/1386595 [00:03<00:00, 620689.82 examples/s]
Saving the d
+1: ataset (59/192 shards):  79%|███████▉  | 1093098/1386595 [00:03<00:00, 620689.82 examples/s]
Saving the dataset (60/192 shards):  79%|███████▉  | 1098320/1386595 [00:03<00:00, 620689.82 examples/s]
Saving the dataset (61/192 shards):  80%|███████▉  | 1104542/1386595 [00:03<00:00, 620689.82 examples/s]
Saving the dataset (61/192 shards):  80%|███████▉  | 1105542/1386595 [00:03<00:00, 621316.48 examples/s]
Saving the dataset (62/192 shards):  80%|███████▉  | 1107764/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (63/192 shards):  80%|████████  | 1109986/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (64/192 shards):  80%|████████  | 1116208/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (65/192 shards):  82%|████████▏ | 1131430/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (66/192 shards):  82%|████████�
+1: �� | 1132874/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (67/192 shards):  82%|████████▏ | 1132874/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (68/192 shards):  82%|████████▏ | 1135096/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (69/192 shards):  82%|████████▏ | 1136318/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (70/192 shards):  82%|████████▏ | 1136540/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (71/192 shards):  82%|████████▏ | 1136762/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (72/192 shards):  82%|████████▏ | 1140984/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (73/192 shards):  82%|████████▏ | 1141428/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (74/192 shards):  82%|████████▏ | 1141428/1386595 [00:04<00:00, 62131
+1: 6.48 examples/s]
Saving the dataset (75/192 shards):  83%|████████▎ | 1144872/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (76/192 shards):  83%|████████▎ | 1144872/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (77/192 shards):  83%|████████▎ | 1149760/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (78/192 shards):  83%|████████▎ | 1150760/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (79/192 shards):  83%|████████▎ | 1150760/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (80/192 shards):  83%|████████▎ | 1151982/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (81/192 shards):  83%|████████▎ | 1152204/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (82/192 shards):  83%|████████▎ | 1152204/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (83/
+1: 192 shards):  83%|████████▎ | 1154426/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (84/192 shards):  83%|████████▎ | 1156648/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (85/192 shards):  84%|████████▍ | 1162870/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (86/192 shards):  84%|████████▍ | 1163092/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (87/192 shards):  84%|████████▍ | 1164314/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (88/192 shards):  84%|████████▍ | 1166536/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (89/192 shards):  85%|████████▍ | 1173758/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (90/192 shards):  85%|████████▍ | 1177980/1386595 [00:04<00:00, 621316.48 examples/s]
Saving the dataset (90/192 shards):  85%|███████�
+1: ��▌ | 1178980/1386595 [00:04<00:00, 595192.72 examples/s]
Saving the dataset (91/192 shards):  85%|████████▌ | 1183202/1386595 [00:04<00:00, 595192.72 examples/s]
Saving the dataset (92/192 shards):  85%|████████▌ | 1184646/1386595 [00:04<00:00, 595192.72 examples/s]
Saving the dataset (93/192 shards):  85%|████████▌ | 1184646/1386595 [00:04<00:00, 595192.72 examples/s]
Saving the dataset (94/192 shards):  86%|████████▌ | 1193868/1386595 [00:04<00:00, 595192.72 examples/s]
Saving the dataset (95/192 shards):  87%|████████▋ | 1206090/1386595 [00:04<00:00, 595192.72 examples/s]
Saving the dataset (96/192 shards):  87%|████████▋ | 1207312/1386595 [00:04<00:00, 595192.72 examples/s]
Saving the dataset (97/192 shards):  87%|████████▋ | 1211534/1386595 [00:04<00:00, 595192.72 examples/s]
Saving the dataset (98/192 shards):  88%|████████▊ | 1214756/1386595 [00:04<00:00, 59
+1: 5192.72 examples/s]
Saving the dataset (99/192 shards):  88%|████████▊ | 1214978/1386595 [00:04<00:00, 595192.72 examples/s]
Saving the dataset (100/192 shards):  88%|████████▊ | 1224644/1386595 [00:04<00:00, 595192.72 examples/s]
Saving the dataset (101/192 shards):  88%|████████▊ | 1225644/1386595 [00:04<00:00, 595192.72 examples/s]
Saving the dataset (102/192 shards):  89%|████████▊ | 1227644/1386595 [00:04<00:00, 595192.72 examples/s]
Saving the dataset (103/192 shards):  89%|████████▊ | 1229088/1386595 [00:04<00:00, 595192.72 examples/s]
Saving the dataset (104/192 shards):  89%|████████▊ | 1229088/1386595 [00:04<00:00, 595192.72 examples/s]
Saving the dataset (105/192 shards):  89%|████████▉ | 1233310/1386595 [00:04<00:00, 595192.72 examples/s]
Saving the dataset (106/192 shards):  89%|████████▉ | 1237532/1386595 [00:04<00:00, 595192.72 examples/s]
Saving the da
+1: taset (107/192 shards):  89%|████████▉ | 1239754/1386595 [00:04<00:00, 595192.72 examples/s]
Saving the dataset (108/192 shards):  89%|████████▉ | 1240976/1386595 [00:04<00:00, 595192.72 examples/s]
Saving the dataset (109/192 shards):  90%|████████▉ | 1245198/1386595 [00:04<00:00, 595192.72 examples/s]
Saving the dataset (110/192 shards):  90%|████████▉ | 1247642/1386595 [00:04<00:00, 595192.72 examples/s]
Saving the dataset (111/192 shards):  90%|████████▉ | 1247642/1386595 [00:04<00:00, 595192.72 examples/s]
Saving the dataset (112/192 shards):  90%|████████▉ | 1247864/1386595 [00:04<00:00, 595192.72 examples/s]
Saving the dataset (113/192 shards):  90%|█████████ | 1248086/1386595 [00:04<00:00, 595192.72 examples/s]
Saving the dataset (113/192 shards):  90%|█████████ | 1249308/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (114/192 shards):  90%|█
+1: ████████ | 1249308/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (115/192 shards):  90%|█████████ | 1249530/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (116/192 shards):  90%|█████████ | 1251752/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (117/192 shards):  90%|█████████ | 1252974/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (118/192 shards):  90%|█████████ | 1253196/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (119/192 shards):  91%|█████████ | 1255418/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (120/192 shards):  91%|█████████ | 1256640/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (121/192 shards):  91%|█████████ | 1258862/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (122/192 shards):  91%|█████████ | 12590
+1: 84/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (123/192 shards):  91%|█████████ | 1259528/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (124/192 shards):  91%|█████████ | 1259528/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (125/192 shards):  91%|█████████ | 1262750/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (126/192 shards):  91%|█████████▏| 1266972/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (127/192 shards):  92%|█████████▏| 1278194/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (128/192 shards):  92%|█████████▏| 1278416/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (129/192 shards):  93%|█████████▎| 1287638/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (130/192 shards):  93%|█████████▎| 1292860/1386595 [00:04<00:0
+1: 0, 618066.21 examples/s]
Saving the dataset (131/192 shards):  94%|█████████▎| 1297304/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (132/192 shards):  94%|█████████▎| 1299304/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (133/192 shards):  94%|█████████▎| 1299748/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (134/192 shards):  94%|█████████▎| 1299748/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (135/192 shards):  94%|█████████▍| 1301969/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (136/192 shards):  94%|█████████▍| 1309191/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (137/192 shards):  94%|█████████▍| 1309635/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (138/192 shards):  94%|█████████▍| 1309635/1386595 [00:04<00:00, 618066.21 exa
+1: mples/s]
Saving the dataset (139/192 shards):  95%|█████████▍| 1311079/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (140/192 shards):  95%|█████████▍| 1311079/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (141/192 shards):  95%|█████████▍| 1315300/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (142/192 shards):  95%|█████████▍| 1315522/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (143/192 shards):  95%|█████████▍| 1315965/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (144/192 shards):  95%|█████████▍| 1315965/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (145/192 shards):  95%|█████████▍| 1316187/1386595 [00:04<00:00, 618066.21 examples/s]
Saving the dataset (145/192 shards):  95%|█████████▌| 1318187/1386595 [00:04<00:00, 607715.29 examples/s]
Saving 
+1: the dataset (146/192 shards):  95%|█████████▌| 1319409/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (147/192 shards):  95%|█████████▌| 1322631/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (148/192 shards):  95%|█████████▌| 1322853/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (149/192 shards):  96%|█████████▌| 1325296/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (150/192 shards):  96%|█████████▌| 1325296/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (151/192 shards):  96%|█████████▌| 1329518/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (152/192 shards):  96%|█████████▌| 1332740/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (153/192 shards):  96%|█████████▌| 1334183/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (154
+1: /192 shards):  96%|█████████▌| 1334183/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (155/192 shards):  96%|█████████▋| 1337404/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (156/192 shards):  97%|█████████▋| 1343848/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (157/192 shards):  97%|█████████▋| 1343848/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (158/192 shards):  97%|█████████▋| 1347069/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (159/192 shards):  97%|█████████▋| 1347290/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (160/192 shards):  97%|█████████▋| 1351512/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (161/192 shards):  97%|█████████▋| 1351734/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (162/192 shards):  9
+1: 8%|█████████▊| 1354177/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (163/192 shards):  98%|█████████▊| 1354177/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (164/192 shards):  98%|█████████▊| 1354399/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (165/192 shards):  98%|█████████▊| 1354842/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (166/192 shards):  98%|█████████▊| 1354842/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (167/192 shards):  98%|█████████▊| 1355063/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (168/192 shards):  98%|█████████▊| 1355284/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (169/192 shards):  98%|█████████▊| 1357505/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (170/192 shards):  98%|████�
+1: ��████▊| 1358727/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (171/192 shards):  98%|█████████▊| 1361948/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (172/192 shards):  98%|█████████▊| 1363170/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (173/192 shards):  98%|█████████▊| 1363392/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (174/192 shards):  98%|█████████▊| 1364613/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (175/192 shards):  99%|█████████▊| 1366056/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (176/192 shards):  99%|█████████▊| 1366056/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (177/192 shards):  99%|█████████▊| 1366278/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (178/192 shards):  99%|█████████�
+1: �| 1367499/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (179/192 shards):  99%|█████████▊| 1367720/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (180/192 shards):  99%|█████████▉| 1373941/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (181/192 shards):  99%|█████████▉| 1375162/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (182/192 shards):  99%|█████████▉| 1377383/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (183/192 shards):  99%|█████████▉| 1377605/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (184/192 shards):  99%|█████████▉| 1377826/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (185/192 shards):  99%|█████████▉| 1379047/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (186/192 shards):  99%|█████████▉| 1379268/13865
+1: 95 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (187/192 shards): 100%|█████████▉| 1381490/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (188/192 shards): 100%|█████████▉| 1381711/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (189/192 shards): 100%|█████████▉| 1382932/1386595 [00:04<00:00, 607715.29 examples/s]
Saving the dataset (189/192 shards): 100%|█████████▉| 1384153/1386595 [00:04<00:00, 531422.00 examples/s]
Saving the dataset (190/192 shards): 100%|█████████▉| 1384153/1386595 [00:04<00:00, 531422.00 examples/s]
Saving the dataset (191/192 shards): 100%|█████████▉| 1385374/1386595 [00:04<00:00, 531422.00 examples/s]
Saving the dataset (192/192 shards): 100%|██████████| 1386595/1386595 [00:04<00:00, 531422.00 examples/s]
Saving the dataset (192/192 shards): 100%|██████████| 1386595/1386595 [00:04<00:00,
+1:  300657.57 examples/s]
+0: [2025-08-06 23:22:58,634] [INFO] [axolotl.utils.data.shared.load_preprocessed_dataset:471] [PID:411490] [RANK:0] Loading prepared dataset from disk at /lustre/fsn1/projects/rech/dgo/udv55np/dataset/Qwen3-235B-A22B/Qwen2.5-0.5B/1/573f14bfb0bdbca74738744c91f193e0...
+0: [2025-08-06 23:27:07,197] [INFO] [axolotl.utils.samplers.multipack.calc_min_len:435] [PID:411490] [RANK:0] gather_len_batches: [235608, 235604, 235607, 235612]
+0: [2025-08-06 23:27:07,246] [INFO] [axolotl.utils.trainer.calc_sample_packing_eff_est:496] [PID:411490] [RANK:0] sample_packing_eff_est across ranks: [0.9936306476593018, 0.9936643838882446, 0.9936602115631104, 0.9936854839324951]
+0: [2025-08-06 23:27:07,270] [INFO] [axolotl.utils.data.sft._prepare_standard_dataset:123] [PID:411490] [RANK:0] Maximum number of steps set at 14725
+2: You are attempting to use Flash Attention 2.0 with a model not initialized on GPU. Make sure to move the model to GPU after initializing it on CPU with `model.to('cuda')`.
+1: You are attempting to use Flash Attention 2.0 with a model not initialized on GPU. Make sure to move the model to GPU after initializing it on CPU with `model.to('cuda')`.
+0: You are attempting to use Flash Attention 2.0 with a model not initialized on GPU. Make sure to move the model to GPU after initializing it on CPU with `model.to('cuda')`.
+3: You are attempting to use Flash Attention 2.0 with a model not initialized on GPU. Make sure to move the model to GPU after initializing it on CPU with `model.to('cuda')`.
+0: [2025-08-06 23:27:21,599] [INFO] [axolotl.loaders.model._configure_embedding_dtypes:317] [PID:411490] [RANK:0] Converting modules to torch.bfloat16
+0: [2025-08-06 23:27:39,958] [INFO] [axolotl.train.save_initial_configs:397] [PID:411490] [RANK:0] Pre-saving tokenizer to /lustre/fswork/projects/rech/dgo/udv55np/ift/Qwen3-235B-A22B/Qwen2.5-0.5B/1...
+0: [2025-08-06 23:27:40,182] [INFO] [axolotl.train.save_initial_configs:400] [PID:411490] [RANK:0] Pre-saving model config to /lustre/fswork/projects/rech/dgo/udv55np/ift/Qwen3-235B-A22B/Qwen2.5-0.5B/1...
+0: [2025-08-06 23:27:40,209] [INFO] [axolotl.train.execute_training:221] [PID:411490] [RANK:0] Starting trainer...
+0: [2025-08-06 23:48:13,983] [INFO] [axolotl.utils.samplers.multipack.calc_min_len:435] [PID:411490] [RANK:0] gather_len_batches: [235606, 235606, 235606, 235606]
+0: Parameter Offload - Persistent parameters statistics: param_count = 121, numel = 71552
+0: {'loss': 1.3643, 'grad_norm': 2.675846961890176, 'learning_rate': 3.08e-06, 'epoch': 0.0}
+0: 
  0%|          | 0/14725 [00:00<?, ?it/s]
  0%|          | 1/14725 [01:04<263:36:58, 64.45s/it]
  0%|          | 2/14725 [01:06<113:00:50, 27.63s/it]
  0%|          | 3/14725 [01:08<64:42:32, 15.82s/it] 
  0%|          | 4/14725 [01:09<42:05:18, 10.29s/it]
  0%|          | 5/14725 [01:11<29:36:32,  7.24s/it]
  0%|          | 6/14725 [01:13<22:03:52,  5.40s/it]
  0%|          | 7/14725 [01:15<17:14:47,  4.22s/it]
  0%|          | 8/14725 [01:17<14:09:08,  3.46s/it]
  0%|          | 9/14725 [01:18<11:59:17,  2.93s/it]
  0%|          | 10/14725 [01:20<10:30:46,  2.57s/it]
                                                     

  0%|          | 10/14725 [01:20<10:30:46,  2.57s/it]
  0%|          | 11/14725 [01:22<9:44:01,  2.38s/it] 
  0%|          | 12/14725 [01:24<8:58:04,  2.19s/it]
  0%|          | 13/14725 [01:26<8:25:34,  2.06s/it]
  0%|          | 14/14725 [01:27<8:03:58,  1.97s/it]
  0%|          | 15/14725 [01:29<7:59:45,  1.96s/it]
  0%|          | 16/14725 [01:31<7:47:37,  1.91s/it]
  0%|          | 17
+0: {'loss': 1.2644, 'grad_norm': 1.6110014238427781, 'learning_rate': 4.2800000000000005e-06, 'epoch': 0.0}
+0: {'loss': 1.2048, 'grad_norm': 1.0036379513023321, 'learning_rate': 5.480000000000001e-06, 'epoch': 0.0}
+0: /14725 [01:33<7:35:47,  1.86s/it]
  0%|          | 18/14725 [01:35<7:38:58,  1.87s/it]
  0%|          | 19/14725 [01:37<7:38:33,  1.87s/it]
  0%|          | 20/14725 [01:39<7:35:52,  1.86s/it]
                                                    

  0%|          | 20/14725 [01:39<7:35:52,  1.86s/it]
  0%|          | 21/14725 [01:40<7:38:00,  1.87s/it]
  0%|          | 22/14725 [01:42<7:34:16,  1.85s/it]
  0%|          | 23/14725 [01:44<7:31:35,  1.84s/it]
  0%|          | 24/14725 [01:46<7:29:06,  1.83s/it]
  0%|          | 25/14725 [01:48<7:23:36,  1.81s/it]
  0%|          | 26/14725 [01:49<7:25:07,  1.82s/it]
  0%|          | 27/14725 [01:51<7:22:31,  1.81s/it]
  0%|          | 28/14725 [01:53<7:25:24,  1.82s/it]
  0%|          | 29/14725 [01:55<7:28:13,  1.83s/it]
  0%|          | 30/14725 [01:57<7:27:56,  1.83s/it]
                                                    

  0%|          | 30/14725 [01:57<7:27:56,  1.83s/it]
  0%|          | 31/14725 [01:59<7:29:00,  1.83s/it]
  0%|          | 32/14725 [02:00<7
+0: {'loss': 1.1661, 'grad_norm': 1.0097042767921711, 'learning_rate': 6.680000000000001e-06, 'epoch': 0.0}
+0: :31:43,  1.84s/it]
  0%|          | 33/14725 [02:02<7:26:54,  1.83s/it]
  0%|          | 34/14725 [02:04<7:26:20,  1.82s/it]
  0%|          | 35/14725 [02:06<7:21:45,  1.80s/it]
  0%|          | 36/14725 [02:08<7:23:38,  1.81s/it]
  0%|          | 37/14725 [02:09<7:24:53,  1.82s/it]
  0%|          | 38/14725 [02:11<7:24:38,  1.82s/it]
  0%|          | 39/14725 [02:13<7:28:08,  1.83s/it]
  0%|          | 40/14725 [02:15<7:27:16,  1.83s/it]
                                                    

  0%|          | 40/14725 [02:15<7:27:16,  1.83s/it]
  0%|          | 41/14725 [02:17<7:31:23,  1.84s/it]
  0%|          | 42/14725 [02:19<7:32:26,  1.85s/it]
  0%|          | 43/14725 [02:21<7:27:31,  1.83s/it]
  0%|          | 44/14725 [02:22<7:23:46,  1.81s/it]
  0%|          | 45/14725 [02:24<7:25:22,  1.82s/it]
  0%|          | 46/14725 [02:26<7:22:41,  1.81s/it]
  0%|          | 47/14725 [02:28<7:21:28,  1.80s/it]
  0%|          | 48/14725 [02:29<7:19:26,  1.80s/it]
  0%|          | 49/14725 [02:31<7:23:35,  1.81s/i
+0: {'loss': 1.1276, 'grad_norm': 0.9234890312182751, 'learning_rate': 7.88e-06, 'epoch': 0.0}
+0: {'loss': 1.1005, 'grad_norm': 0.9731757092865839, 'learning_rate': 9.08e-06, 'epoch': 0.0}
+0: t]
  0%|          | 50/14725 [02:33<7:22:35,  1.81s/it]
                                                    

  0%|          | 50/14725 [02:33<7:22:35,  1.81s/it]
  0%|          | 51/14725 [02:35<7:27:14,  1.83s/it]
  0%|          | 52/14725 [02:37<7:24:50,  1.82s/it]
  0%|          | 53/14725 [02:39<7:22:37,  1.81s/it]
  0%|          | 54/14725 [02:40<7:27:56,  1.83s/it]
  0%|          | 55/14725 [02:42<7:29:47,  1.84s/it]
  0%|          | 56/14725 [02:44<7:29:47,  1.84s/it]
  0%|          | 57/14725 [02:46<7:23:33,  1.81s/it]
  0%|          | 58/14725 [02:48<7:18:56,  1.80s/it]
  0%|          | 59/14725 [02:49<7:18:19,  1.79s/it]
  0%|          | 60/14725 [02:51<7:19:22,  1.80s/it]
                                                    

  0%|          | 60/14725 [02:51<7:19:22,  1.80s/it]
  0%|          | 61/14725 [02:53<7:23:59,  1.82s/it]
  0%|          | 62/14725 [02:55<7:19:49,  1.80s/it]
  0%|          | 63/14725 [02:57<7:19:19,  1.80s/it]
  0%|          | 64/14725 [02:59<7:24:50,  1.82s/it]
  0%|       
+0: {'loss': 1.0936, 'grad_norm': 0.9296007982444164, 'learning_rate': 1.0280000000000002e-05, 'epoch': 0.0}
+0: {'loss': 1.0784, 'grad_norm': 0.9778966592380071, 'learning_rate': 1.148e-05, 'epoch': 0.01}
+0:    | 65/14725 [03:00<7:22:14,  1.81s/it]
  0%|          | 66/14725 [03:02<7:22:00,  1.81s/it]
  0%|          | 67/14725 [03:04<7:22:04,  1.81s/it]
  0%|          | 68/14725 [03:06<7:19:26,  1.80s/it]
  0%|          | 69/14725 [03:08<7:18:33,  1.80s/it]
  0%|          | 70/14725 [03:09<7:19:05,  1.80s/it]
                                                    

  0%|          | 70/14725 [03:09<7:19:05,  1.80s/it]
  0%|          | 71/14725 [03:11<7:32:18,  1.85s/it]
  0%|          | 72/14725 [03:13<7:27:12,  1.83s/it]
  0%|          | 73/14725 [03:15<7:25:02,  1.82s/it]
  1%|          | 74/14725 [03:17<7:21:12,  1.81s/it]
  1%|          | 75/14725 [03:18<7:21:48,  1.81s/it]
  1%|          | 76/14725 [03:20<7:23:55,  1.82s/it]
  1%|          | 77/14725 [03:22<7:21:19,  1.81s/it]
  1%|          | 78/14725 [03:24<7:21:20,  1.81s/it]
  1%|          | 79/14725 [03:26<7:20:08,  1.80s/it]
  1%|          | 80/14725 [03:28<7:22:09,  1.81s/it]
                                                    

  1%|          | 80/14725 [
+0: {'loss': 1.0779, 'grad_norm': 0.9188856456423758, 'learning_rate': 1.268e-05, 'epoch': 0.01}
+0: 03:28<7:22:09,  1.81s/it]
  1%|          | 81/14725 [03:29<7:19:30,  1.80s/it]
  1%|          | 82/14725 [03:31<7:17:07,  1.79s/it]
  1%|          | 83/14725 [03:33<7:24:20,  1.82s/it]
  1%|          | 84/14725 [03:35<7:22:50,  1.81s/it]
  1%|          | 85/14725 [03:37<7:23:30,  1.82s/it]
  1%|          | 86/14725 [03:39<7:35:57,  1.87s/it]
  1%|          | 87/14725 [03:40<7:28:17,  1.84s/it]
  1%|          | 88/14725 [03:42<7:24:41,  1.82s/it]
  1%|          | 89/14725 [03:44<7:28:11,  1.84s/it]
  1%|          | 90/14725 [03:46<7:25:32,  1.83s/it]
                                                    

  1%|          | 90/14725 [03:46<7:25:32,  1.83s/it]
  1%|          | 91/14725 [03:48<7:24:55,  1.82s/it]
  1%|          | 92/14725 [03:49<7:21:53,  1.81s/it]
  1%|          | 93/14725 [03:51<7:26:59,  1.83s/it]
  1%|          | 94/14725 [03:53<7:27:06,  1.83s/it]
  1%|          | 95/14725 [03:55<7:21:08,  1.81s/it]
  1%|          | 96/14725 [03:57<7:21:41,  1.81s/it]
  1%|          | 97/14725 [03:58<7:20:10,  
+0: {'loss': 1.0582, 'grad_norm': 1.1214704852722133, 'learning_rate': 1.3880000000000003e-05, 'epoch': 0.01}
+0: {'loss': 1.0258, 'grad_norm': 1.074729707065513, 'learning_rate': 1.5080000000000001e-05, 'epoch': 0.01}
+0: 1.81s/it]
  1%|          | 98/14725 [04:00<7:22:32,  1.82s/it]
  1%|          | 99/14725 [04:02<7:24:59,  1.83s/it]
  1%|          | 100/14725 [04:04<7:24:09,  1.82s/it]
                                                     

  1%|          | 100/14725 [04:04<7:24:09,  1.82s/it]
  1%|          | 101/14725 [04:06<7:21:12,  1.81s/it]
  1%|          | 102/14725 [04:08<7:21:56,  1.81s/it]
  1%|          | 103/14725 [04:09<7:26:23,  1.83s/it]
  1%|          | 104/14725 [04:11<7:24:39,  1.82s/it]
  1%|          | 105/14725 [04:13<7:21:24,  1.81s/it]
  1%|          | 106/14725 [04:15<7:22:12,  1.81s/it]
  1%|          | 107/14725 [04:17<7:25:39,  1.83s/it]
  1%|          | 108/14725 [04:19<7:22:06,  1.81s/it]
  1%|          | 109/14725 [04:20<7:24:30,  1.82s/it]
  1%|          | 110/14725 [04:22<7:21:00,  1.81s/it]
                                                     

  1%|          | 110/14725 [04:22<7:21:00,  1.81s/it]
  1%|          | 111/14725 [04:24<7:17:16,  1.80s/it]
  1%|          | 112/14725 [04:26<7:19:26,
+0: {'loss': 1.0274, 'grad_norm': 1.0677030411607875, 'learning_rate': 1.628e-05, 'epoch': 0.01}
+0:   1.80s/it]
  1%|          | 113/14725 [04:28<7:22:48,  1.82s/it]
  1%|          | 114/14725 [04:29<7:20:27,  1.81s/it]
  1%|          | 115/14725 [04:31<7:21:09,  1.81s/it]
  1%|          | 116/14725 [04:33<7:19:10,  1.80s/it]
  1%|          | 117/14725 [04:35<7:17:30,  1.80s/it]
  1%|          | 118/14725 [04:37<7:20:50,  1.81s/it]
  1%|          | 119/14725 [04:38<7:17:22,  1.80s/it]
  1%|          | 120/14725 [04:40<7:20:07,  1.81s/it]
                                                     

  1%|          | 120/14725 [04:40<7:20:07,  1.81s/it]
  1%|          | 121/14725 [04:42<7:22:49,  1.82s/it]
  1%|          | 122/14725 [04:44<7:21:42,  1.81s/it]
  1%|          | 123/14725 [04:46<7:18:14,  1.80s/it]
  1%|          | 124/14725 [04:48<7:28:14,  1.84s/it]
  1%|          | 125/14725 [04:49<7:24:14,  1.83s/it]
  1%|          | 126/14725 [04:51<7:23:08,  1.82s/it]
  1%|          | 127/14725 [04:53<7:23:33,  1.82s/it]
  1%|          | 128/14725 [04:55<7:23:49,  1.82s/it]
  1%|          | 129/14725 [04:57<7:20:
+0: {'loss': 1.0515, 'grad_norm': 1.079897049487233, 'learning_rate': 1.7480000000000002e-05, 'epoch': 0.01}
+0: {'loss': 1.0136, 'grad_norm': 1.1232807744475277, 'learning_rate': 1.868e-05, 'epoch': 0.01}
+0: 29,  1.81s/it]
  1%|          | 130/14725 [04:58<7:16:28,  1.79s/it]
                                                     

  1%|          | 130/14725 [04:58<7:16:28,  1.79s/it]
  1%|          | 131/14725 [05:00<7:22:19,  1.82s/it]
  1%|          | 132/14725 [05:02<7:23:33,  1.82s/it]
  1%|          | 133/14725 [05:04<7:19:18,  1.81s/it]
  1%|          | 134/14725 [05:06<7:28:22,  1.84s/it]
  1%|          | 135/14725 [05:08<7:29:41,  1.85s/it]
  1%|          | 136/14725 [05:09<7:28:31,  1.84s/it]
  1%|          | 137/14725 [05:11<7:26:40,  1.84s/it]
  1%|          | 138/14725 [05:13<7:23:45,  1.83s/it]
  1%|          | 139/14725 [05:15<7:19:27,  1.81s/it]
  1%|          | 140/14725 [05:17<7:17:06,  1.80s/it]
                                                     

  1%|          | 140/14725 [05:17<7:17:06,  1.80s/it]
  1%|          | 141/14725 [05:18<7:14:40,  1.79s/it]
  1%|          | 142/14725 [05:20<7:15:03,  1.79s/it]
  1%|          | 143/14725 [05:22<7:18:46,  1.81s/it]
  1%|          | 144/14725 [05:24<7
+0: {'loss': 1.0246, 'grad_norm': 1.0712840374747257, 'learning_rate': 1.9880000000000003e-05, 'epoch': 0.01}
+0: :20:43,  1.81s/it]
  1%|          | 145/14725 [05:26<7:23:32,  1.83s/it]
  1%|          | 146/14725 [05:28<7:23:39,  1.83s/it]
  1%|          | 147/14725 [05:29<7:25:40,  1.83s/it]
  1%|          | 148/14725 [05:31<7:21:52,  1.82s/it]
  1%|          | 149/14725 [05:33<7:20:16,  1.81s/it]
  1%|          | 150/14725 [05:35<7:17:23,  1.80s/it]
                                                     

  1%|          | 150/14725 [05:35<7:17:23,  1.80s/it]
  1%|          | 151/14725 [05:37<7:22:19,  1.82s/it]
  1%|          | 152/14725 [05:38<7:27:03,  1.84s/it]
  1%|          | 153/14725 [05:40<7:26:05,  1.84s/it]
  1%|          | 154/14725 [05:42<7:23:13,  1.83s/it]
  1%|          | 155/14725 [05:44<7:23:50,  1.83s/it]
  1%|          | 156/14725 [05:46<7:20:16,  1.81s/it]
  1%|          | 157/14725 [05:48<7:22:30,  1.82s/it]
  1%|          | 158/14725 [05:50<7:33:38,  1.87s/it]
  1%|          | 159/14725 [05:51<7:37:40,  1.89s/it]
  1%|          | 160/14725 [05:53<7:30:10,  1.85s/it]
                                
+0: {'loss': 1.0066, 'grad_norm': 1.0863116273859414, 'learning_rate': 2e-05, 'epoch': 0.01}
+0: {'loss': 1.0146, 'grad_norm': 1.1136956695879134, 'learning_rate': 2e-05, 'epoch': 0.01}
+0:                      

  1%|          | 160/14725 [05:53<7:30:10,  1.85s/it]
  1%|          | 161/14725 [05:55<7:28:00,  1.85s/it]
  1%|          | 162/14725 [05:57<7:25:22,  1.83s/it]
  1%|          | 163/14725 [05:59<7:27:30,  1.84s/it]
  1%|          | 164/14725 [06:01<7:25:53,  1.84s/it]
  1%|          | 165/14725 [06:02<7:24:18,  1.83s/it]
  1%|          | 166/14725 [06:04<7:25:49,  1.84s/it]
  1%|          | 167/14725 [06:06<7:30:32,  1.86s/it]
  1%|          | 168/14725 [06:08<7:28:23,  1.85s/it]
  1%|          | 169/14725 [06:10<7:23:19,  1.83s/it]
  1%|          | 170/14725 [06:12<7:23:57,  1.83s/it]
                                                     

  1%|          | 170/14725 [06:12<7:23:57,  1.83s/it]
  1%|          | 171/14725 [06:13<7:21:31,  1.82s/it]
  1%|          | 172/14725 [06:15<7:21:09,  1.82s/it]
  1%|          | 173/14725 [06:17<7:19:06,  1.81s/it]
  1%|          | 174/14725 [06:19<7:18:58,  1.81s/it]
  1%|          | 175/14725 [06:21<7:20:11,  1.82s/it]
  1%|          | 176/14725 [
+0: {'loss': 1.0044, 'grad_norm': 1.124505543393061, 'learning_rate': 2e-05, 'epoch': 0.01}
+0: {'loss': 0.9909, 'grad_norm': 1.263280138300491, 'learning_rate': 2e-05, 'epoch': 0.01}
+0: 06:22<7:19:02,  1.81s/it]
  1%|          | 177/14725 [06:24<7:23:49,  1.83s/it]
  1%|          | 178/14725 [06:26<7:23:43,  1.83s/it]
  1%|          | 179/14725 [06:28<7:22:27,  1.83s/it]
  1%|          | 180/14725 [06:30<7:31:10,  1.86s/it]
                                                     

  1%|          | 180/14725 [06:30<7:31:10,  1.86s/it]
  1%|          | 181/14725 [06:32<7:26:35,  1.84s/it]
  1%|          | 182/14725 [06:33<7:22:53,  1.83s/it]
  1%|          | 183/14725 [06:35<7:22:05,  1.82s/it]
  1%|          | 184/14725 [06:37<7:20:29,  1.82s/it]
  1%|▏         | 185/14725 [06:39<7:16:30,  1.80s/it]
  1%|▏         | 186/14725 [06:41<7:18:50,  1.81s/it]
  1%|▏         | 187/14725 [06:42<7:17:26,  1.81s/it]
  1%|▏         | 188/14725 [06:44<7:16:43,  1.80s/it]
  1%|▏         | 189/14725 [06:46<7:14:45,  1.79s/it]
  1%|▏         | 190/14725 [06:48<7:13:55,  1.79s/it]
                                                     

  1%|▏         | 190/14725 [06:48<7:13:55,  1.79s/it]
  1%|▏  
+0: {'loss': 1.0155, 'grad_norm': 1.1239379297223577, 'learning_rate': 2e-05, 'epoch': 0.01}
+0:        | 191/14725 [06:50<7:17:14,  1.81s/it]
  1%|▏         | 192/14725 [06:52<7:19:57,  1.82s/it]
  1%|▏         | 193/14725 [06:53<7:17:50,  1.81s/it]
  1%|▏         | 194/14725 [06:55<7:18:47,  1.81s/it]
  1%|▏         | 195/14725 [06:57<7:21:45,  1.82s/it]
  1%|▏         | 196/14725 [06:59<7:23:37,  1.83s/it]
  1%|▏         | 197/14725 [07:01<7:22:18,  1.83s/it]
  1%|▏         | 198/14725 [07:02<7:19:31,  1.82s/it]
  1%|▏         | 199/14725 [07:04<7:21:26,  1.82s/it]
  1%|▏         | 200/14725 [07:06<7:26:57,  1.85s/it]
                                                     

  1%|▏         | 200/14725 [07:06<7:26:57,  1.85s/it]
  1%|▏         | 201/14725 [07:08<7:22:04,  1.83s/it]
  1%|▏         | 202/14725 [07:10<7:24:59,  1.84s/it]
  1%|▏         | 203/14725 [07:12<7:25:18,  1.84s/it]
  1%|▏         | 204/14725 [07:13<7:23:21,  1.83s/it]
  1%|▏         | 205/14725 [07:15<7:19:22,  1.82s/it]
  1%|▏         | 206/14725 [07:17<7:21:03,  1.82s/it]
  1%|▏         | 207/1472
+0: {'loss': 0.9882, 'grad_norm': 1.0132637316433235, 'learning_rate': 2e-05, 'epoch': 0.01}
+0: {'loss': 1.0062, 'grad_norm': 1.022752388877242, 'learning_rate': 2e-05, 'epoch': 0.01}
+0: 5 [07:19<7:16:38,  1.80s/it]
  1%|▏         | 208/14725 [07:21<7:20:22,  1.82s/it]
  1%|▏         | 209/14725 [07:23<7:21:21,  1.82s/it]
  1%|▏         | 210/14725 [07:24<7:19:43,  1.82s/it]
                                                     

  1%|▏         | 210/14725 [07:24<7:19:43,  1.82s/it]
  1%|▏         | 211/14725 [07:26<7:16:18,  1.80s/it]
  1%|▏         | 212/14725 [07:28<7:16:07,  1.80s/it]
  1%|▏         | 213/14725 [07:30<7:14:18,  1.80s/it]
  1%|▏         | 214/14725 [07:31<7:13:29,  1.79s/it]
  1%|▏         | 215/14725 [07:33<7:15:55,  1.80s/it]
  1%|▏         | 216/14725 [07:35<7:17:11,  1.81s/it]
  1%|▏         | 217/14725 [07:37<7:15:22,  1.80s/it]
  1%|▏         | 218/14725 [07:39<7:13:06,  1.79s/it]
  1%|▏         | 219/14725 [07:40<7:09:50,  1.78s/it]
  1%|▏         | 220/14725 [07:42<7:12:39,  1.79s/it]
                                                     

  1%|▏         | 220/14725 [07:42<7:12:39,  1.79s/it]
  2%|▏         | 221/14725 [07:44<7:10:29, 
+0: {'loss': 1.0011, 'grad_norm': 1.05530242922146, 'learning_rate': 2e-05, 'epoch': 0.02}
+0:  1.78s/it]
  2%|▏         | 222/14725 [07:46<7:11:01,  1.78s/it]
  2%|▏         | 223/14725 [07:48<7:13:53,  1.80s/it]
  2%|▏         | 224/14725 [07:49<7:14:22,  1.80s/it]
  2%|▏         | 225/14725 [07:51<7:19:44,  1.82s/it]
  2%|▏         | 226/14725 [07:53<7:19:12,  1.82s/it]
  2%|▏         | 227/14725 [07:55<7:15:51,  1.80s/it]
  2%|▏         | 228/14725 [07:57<7:22:16,  1.83s/it]
  2%|▏         | 229/14725 [07:59<7:23:39,  1.84s/it]
  2%|▏         | 230/14725 [08:00<7:24:29,  1.84s/it]
                                                     

  2%|▏         | 230/14725 [08:00<7:24:29,  1.84s/it]
  2%|▏         | 231/14725 [08:02<7:18:46,  1.82s/it]
  2%|▏         | 232/14725 [08:04<7:18:45,  1.82s/it]
  2%|▏         | 233/14725 [08:06<7:28:44,  1.86s/it]
  2%|▏         | 234/14725 [08:08<7:28:02,  1.86s/it]
  2%|▏         | 235/14725 [08:10<7:24:36,  1.84s/it]
  2%|▏         | 236/14725 [08:12<7:25:26,  1.84s/it]
  2%|▏         | 237/14725 [08:13<7:26:47,  1.85s/it]
  2%|�
+0: {'loss': 1.0024, 'grad_norm': 1.0783706911728916, 'learning_rate': 2e-05, 'epoch': 0.02}
+0: {'loss': 1.0009, 'grad_norm': 1.0251928226413354, 'learning_rate': 2e-05, 'epoch': 0.02}
+0: ��         | 238/14725 [08:15<7:27:47,  1.85s/it]
  2%|▏         | 239/14725 [08:17<7:24:31,  1.84s/it]
  2%|▏         | 240/14725 [08:19<7:19:47,  1.82s/it]
                                                     

  2%|▏         | 240/14725 [08:19<7:19:47,  1.82s/it]
  2%|▏         | 241/14725 [08:21<7:21:06,  1.83s/it]
  2%|▏         | 242/14725 [08:22<7:18:29,  1.82s/it]
  2%|▏         | 243/14725 [08:24<7:18:57,  1.82s/it]
  2%|▏         | 244/14725 [08:26<7:23:33,  1.84s/it]
  2%|▏         | 245/14725 [08:28<7:25:40,  1.85s/it]
  2%|▏         | 246/14725 [08:30<7:20:49,  1.83s/it]
  2%|▏         | 247/14725 [08:32<7:19:14,  1.82s/it]
  2%|▏         | 248/14725 [08:33<7:15:37,  1.81s/it]
  2%|▏         | 249/14725 [08:35<7:17:30,  1.81s/it]
  2%|▏         | 250/14725 [08:37<7:18:06,  1.82s/it]
                                                     

  2%|▏         | 250/14725 [08:37<7:18:06,  1.82s/it]
  2%|▏         | 251/14725 [08:39<7:16:54,  1.81s/it]
  2%|▏         | 252/1
+0: {'loss': 0.9851, 'grad_norm': 1.0568357594828475, 'learning_rate': 2e-05, 'epoch': 0.02}
+0: 4725 [08:41<7:15:34,  1.81s/it]
  2%|▏         | 253/14725 [08:42<7:15:04,  1.80s/it]
  2%|▏         | 254/14725 [08:44<7:11:08,  1.79s/it]
  2%|▏         | 255/14725 [08:46<7:12:13,  1.79s/it]
  2%|▏         | 256/14725 [08:48<7:09:57,  1.78s/it]
  2%|▏         | 257/14725 [08:50<7:09:24,  1.78s/it]
  2%|▏         | 258/14725 [08:51<7:18:47,  1.82s/it]
  2%|▏         | 259/14725 [08:53<7:24:29,  1.84s/it]
  2%|▏         | 260/14725 [08:55<7:28:19,  1.86s/it]
                                                     

  2%|▏         | 260/14725 [08:55<7:28:19,  1.86s/it]
  2%|▏         | 261/14725 [08:57<7:29:46,  1.87s/it]
  2%|▏         | 262/14725 [08:59<7:24:42,  1.84s/it]
  2%|▏         | 263/14725 [09:01<7:27:50,  1.86s/it]
  2%|▏         | 264/14725 [09:03<7:27:37,  1.86s/it]
  2%|▏         | 265/14725 [09:04<7:23:54,  1.84s/it]
  2%|▏         | 266/14725 [09:06<7:19:56,  1.83s/it]
  2%|▏         | 267/14725 [09:08<7:15:42,  1.81s/it]
  2%|▏         | 268/14725 [09:10<7:19:
+0: {'loss': 0.9961, 'grad_norm': 1.1373634747708403, 'learning_rate': 2e-05, 'epoch': 0.02}
+0: {'loss': 0.9868, 'grad_norm': 1.0851314337238565, 'learning_rate': 2e-05, 'epoch': 0.02}
+0: 36,  1.82s/it]
  2%|▏         | 269/14725 [09:12<7:22:22,  1.84s/it]
  2%|▏         | 270/14725 [09:14<7:25:42,  1.85s/it]
                                                     

  2%|▏         | 270/14725 [09:14<7:25:42,  1.85s/it]
  2%|▏         | 271/14725 [09:15<7:20:08,  1.83s/it]
  2%|▏         | 272/14725 [09:17<7:17:30,  1.82s/it]
  2%|▏         | 273/14725 [09:19<7:18:37,  1.82s/it]
  2%|▏         | 274/14725 [09:21<7:14:45,  1.81s/it]
  2%|▏         | 275/14725 [09:23<7:12:30,  1.80s/it]
  2%|▏         | 276/14725 [09:24<7:16:12,  1.81s/it]
  2%|▏         | 277/14725 [09:26<7:19:05,  1.82s/it]
  2%|▏         | 278/14725 [09:28<7:16:20,  1.81s/it]
  2%|▏         | 279/14725 [09:30<7:32:40,  1.88s/it]
  2%|▏         | 280/14725 [09:32<7:27:52,  1.86s/it]
                                                     

  2%|▏         | 280/14725 [09:32<7:27:52,  1.86s/it]
  2%|▏         | 281/14725 [09:34<7:25:52,  1.85s/it]
  2%|▏         | 282/14725 [09:36<7:19:30,  1.83s/it]
  2
+0: {'loss': 0.9951, 'grad_norm': 1.0120087617604328, 'learning_rate': 2e-05, 'epoch': 0.02}
+0: %|▏         | 283/14725 [09:37<7:24:24,  1.85s/it]
  2%|▏         | 284/14725 [09:39<7:20:54,  1.83s/it]
  2%|▏         | 285/14725 [09:41<7:17:44,  1.82s/it]
  2%|▏         | 286/14725 [09:43<7:15:38,  1.81s/it]
  2%|▏         | 287/14725 [09:45<7:20:07,  1.83s/it]
  2%|▏         | 288/14725 [09:46<7:19:12,  1.83s/it]
  2%|▏         | 289/14725 [09:48<7:26:17,  1.85s/it]
  2%|▏         | 290/14725 [09:50<7:25:43,  1.85s/it]
                                                     

  2%|▏         | 290/14725 [09:50<7:25:43,  1.85s/it]
  2%|▏         | 291/14725 [09:52<7:22:26,  1.84s/it]
  2%|▏         | 292/14725 [09:54<7:21:53,  1.84s/it]
  2%|▏         | 293/14725 [09:56<7:23:23,  1.84s/it]
  2%|▏         | 294/14725 [09:58<7:24:30,  1.85s/it]
  2%|▏         | 295/14725 [09:59<7:23:54,  1.85s/it]
  2%|▏         | 296/14725 [10:01<7:23:47,  1.85s/it]
  2%|▏         | 297/14725 [10:03<7:23:07,  1.84s/it]
  2%|▏         | 298/14725 [10:05<7:17:32,  1.82s/it]
  2%|▏         | 2
+0: {'loss': 0.987, 'grad_norm': 0.8875862913298864, 'learning_rate': 2e-05, 'epoch': 0.02}
+0: {'loss': 0.9879, 'grad_norm': 1.0500335395029998, 'learning_rate': 2e-05, 'epoch': 0.02}
+0: 99/14725 [10:07<7:14:03,  1.81s/it]
  2%|▏         | 300/14725 [10:08<7:11:36,  1.80s/it]
                                                     

  2%|▏         | 300/14725 [10:08<7:11:36,  1.80s/it]
  2%|▏         | 301/14725 [10:10<7:14:26,  1.81s/it]
  2%|▏         | 302/14725 [10:12<7:12:11,  1.80s/it]
  2%|▏         | 303/14725 [10:14<7:13:11,  1.80s/it]
  2%|▏         | 304/14725 [10:16<7:13:21,  1.80s/it]
  2%|▏         | 305/14725 [10:17<7:11:27,  1.80s/it]
  2%|▏         | 306/14725 [10:19<7:19:20,  1.83s/it]
  2%|▏         | 307/14725 [10:21<7:32:38,  1.88s/it]
  2%|▏         | 308/14725 [10:23<7:25:13,  1.85s/it]
  2%|▏         | 309/14725 [10:25<7:24:26,  1.85s/it]
  2%|▏         | 310/14725 [10:27<7:22:46,  1.84s/it]
                                                     

  2%|▏         | 310/14725 [10:27<7:22:46,  1.84s/it]
  2%|▏         | 311/14725 [10:29<7:20:40,  1.83s/it]
  2%|▏         | 312/14725 [10:30<7:16:44,  1.82s/it]
  2%|▏         | 313/14725 [10:32<7:
+0: {'loss': 0.9556, 'grad_norm': 1.0332762305299081, 'learning_rate': 2e-05, 'epoch': 0.02}
+0: 29:12,  1.87s/it]
  2%|▏         | 314/14725 [10:34<7:27:24,  1.86s/it]
  2%|▏         | 315/14725 [10:36<7:26:28,  1.86s/it]
  2%|▏         | 316/14725 [10:38<7:25:29,  1.86s/it]
  2%|▏         | 317/14725 [10:40<7:23:49,  1.85s/it]
  2%|▏         | 318/14725 [10:42<7:35:41,  1.90s/it]
  2%|▏         | 319/14725 [10:44<7:31:34,  1.88s/it]
  2%|▏         | 320/14725 [10:45<7:27:44,  1.86s/it]
                                                     

  2%|▏         | 320/14725 [10:45<7:27:44,  1.86s/it]
  2%|▏         | 321/14725 [10:47<7:24:31,  1.85s/it]
  2%|▏         | 322/14725 [10:49<7:24:30,  1.85s/it]
  2%|▏         | 323/14725 [10:51<7:27:40,  1.87s/it]
  2%|▏         | 324/14725 [10:53<7:25:19,  1.86s/it]
  2%|▏         | 325/14725 [10:55<7:22:52,  1.85s/it]
  2%|▏         | 326/14725 [10:57<7:26:00,  1.86s/it]
  2%|▏         | 327/14725 [10:58<7:20:34,  1.84s/it]
  2%|▏         | 328/14725 [11:00<7:19:36,  1.83s/it]
  2%|▏         | 329/14725 [11:02<7:20:17,  1.84s/it]
+0: {'loss': 0.9981, 'grad_norm': 0.9658605832329531, 'learning_rate': 2e-05, 'epoch': 0.02}
+0: {'loss': 0.9508, 'grad_norm': 1.0163993977519343, 'learning_rate': 2e-05, 'epoch': 0.02}
+0: 
  2%|▏         | 330/14725 [11:04<7:19:01,  1.83s/it]
                                                     

  2%|▏         | 330/14725 [11:04<7:19:01,  1.83s/it]
  2%|▏         | 331/14725 [11:06<7:19:53,  1.83s/it]
  2%|▏         | 332/14725 [11:07<7:17:36,  1.82s/it]
  2%|▏         | 333/14725 [11:09<7:16:29,  1.82s/it]
  2%|▏         | 334/14725 [11:11<7:19:30,  1.83s/it]
  2%|▏         | 335/14725 [11:13<7:16:57,  1.82s/it]
  2%|▏         | 336/14725 [11:15<7:15:06,  1.81s/it]
  2%|▏         | 337/14725 [11:17<7:13:30,  1.81s/it]
  2%|▏         | 338/14725 [11:18<7:16:13,  1.82s/it]
  2%|▏         | 339/14725 [11:20<7:19:04,  1.83s/it]
  2%|▏         | 340/14725 [11:22<7:24:21,  1.85s/it]
                                                     

  2%|▏         | 340/14725 [11:22<7:24:21,  1.85s/it]
  2%|▏         | 341/14725 [11:24<7:22:42,  1.85s/it]
  2%|▏         | 342/14725 [11:26<7:20:58,  1.84s/it]
  2%|▏         | 343/14725 [11:28<7:17:10,  1.82s/it]
  2%|▏         
+0: {'loss': 0.9749, 'grad_norm': 1.0546490838832316, 'learning_rate': 2e-05, 'epoch': 0.02}
+0: | 344/14725 [11:29<7:15:35,  1.82s/it]
  2%|▏         | 345/14725 [11:31<7:18:12,  1.83s/it]
  2%|▏         | 346/14725 [11:33<7:14:52,  1.81s/it]
  2%|▏         | 347/14725 [11:35<7:17:00,  1.82s/it]
  2%|▏         | 348/14725 [11:37<7:17:46,  1.83s/it]
  2%|▏         | 349/14725 [11:38<7:15:15,  1.82s/it]
  2%|▏         | 350/14725 [11:40<7:19:35,  1.83s/it]
                                                     

  2%|▏         | 350/14725 [11:40<7:19:35,  1.83s/it]
  2%|▏         | 351/14725 [11:42<7:24:12,  1.85s/it]
  2%|▏         | 352/14725 [11:44<7:27:53,  1.87s/it]
  2%|▏         | 353/14725 [11:46<7:22:55,  1.85s/it]
  2%|▏         | 354/14725 [11:48<7:18:01,  1.83s/it]
  2%|▏         | 355/14725 [11:50<7:19:09,  1.83s/it]
  2%|▏         | 356/14725 [11:51<7:19:49,  1.84s/it]
  2%|▏         | 357/14725 [11:53<7:19:52,  1.84s/it]
  2%|▏         | 358/14725 [11:55<7:24:16,  1.86s/it]
  2%|▏         | 359/14725 [11:57<7:31:10,  1.88s/it]
  2%|▏         | 360/14725 [11:5
+0: {'loss': 0.9598, 'grad_norm': 0.8722792202690675, 'learning_rate': 2e-05, 'epoch': 0.02}
+0: {'loss': 0.9631, 'grad_norm': 1.1146986299609514, 'learning_rate': 2e-05, 'epoch': 0.03}
+0: 9<7:26:30,  1.87s/it]
                                                     

  2%|▏         | 360/14725 [11:59<7:26:30,  1.87s/it]
  2%|▏         | 361/14725 [12:01<7:27:59,  1.87s/it]
  2%|▏         | 362/14725 [12:03<7:30:24,  1.88s/it]
  2%|▏         | 363/14725 [12:05<7:27:38,  1.87s/it]
  2%|▏         | 364/14725 [12:06<7:30:39,  1.88s/it]
  2%|▏         | 365/14725 [12:08<7:26:36,  1.87s/it]
  2%|▏         | 366/14725 [12:10<7:28:20,  1.87s/it]
  2%|▏         | 367/14725 [12:12<7:28:22,  1.87s/it]
  2%|▏         | 368/14725 [12:14<7:25:57,  1.86s/it]
  3%|▎         | 369/14725 [12:16<7:27:07,  1.87s/it]
  3%|▎         | 370/14725 [12:18<7:24:27,  1.86s/it]
                                                     

  3%|▎         | 370/14725 [12:18<7:24:27,  1.86s/it]
  3%|▎         | 371/14725 [12:20<7:27:55,  1.87s/it]
  3%|▎         | 372/14725 [12:21<7:27:33,  1.87s/it]
  3%|▎         | 373/14725 [12:23<7:23:48,  1.86s/it]
  3%|▎         | 374/14725 [12:25<7:21:19,  1.85s/
+0: {'loss': 0.9766, 'grad_norm': 1.0278314523620495, 'learning_rate': 2e-05, 'epoch': 0.03}
+0: it]
  3%|▎         | 375/14725 [12:27<7:18:37,  1.83s/it]
  3%|▎         | 376/14725 [12:29<7:20:05,  1.84s/it]
  3%|▎         | 377/14725 [12:31<7:28:23,  1.88s/it]
  3%|▎         | 378/14725 [12:33<7:38:33,  1.92s/it]
  3%|▎         | 379/14725 [12:35<7:31:06,  1.89s/it]
  3%|▎         | 380/14725 [12:36<7:28:45,  1.88s/it]
                                                     

  3%|▎         | 380/14725 [12:36<7:28:45,  1.88s/it]
  3%|▎         | 381/14725 [12:38<7:25:06,  1.86s/it]
  3%|▎         | 382/14725 [12:40<7:22:29,  1.85s/it]
  3%|▎         | 383/14725 [12:42<7:23:16,  1.85s/it]
  3%|▎         | 384/14725 [12:44<7:26:28,  1.87s/it]
  3%|▎         | 385/14725 [12:46<7:25:03,  1.86s/it]
  3%|▎         | 386/14725 [12:47<7:22:12,  1.85s/it]
  3%|▎         | 387/14725 [12:49<7:16:57,  1.83s/it]
  3%|▎         | 388/14725 [12:51<7:19:26,  1.84s/it]
  3%|▎         | 389/14725 [12:53<7:22:06,  1.85s/it]
  3%|▎         | 390/14725 [12:55<7:22:16,  1.85s/it]
             
+0: {'loss': 0.9732, 'grad_norm': 0.913881080861294, 'learning_rate': 2e-05, 'epoch': 0.03}
+0: {'loss': 0.9772, 'grad_norm': 1.2281474909230528, 'learning_rate': 2e-05, 'epoch': 0.03}
+0:                                         

  3%|▎         | 390/14725 [12:55<7:22:16,  1.85s/it]
  3%|▎         | 391/14725 [12:57<7:19:48,  1.84s/it]
  3%|▎         | 392/14725 [12:58<7:20:54,  1.85s/it]
  3%|▎         | 393/14725 [13:00<7:23:33,  1.86s/it]
  3%|▎         | 394/14725 [13:02<7:30:27,  1.89s/it]
  3%|▎         | 395/14725 [13:04<7:26:11,  1.87s/it]
  3%|▎         | 396/14725 [13:06<7:21:16,  1.85s/it]
  3%|▎         | 397/14725 [13:08<7:24:11,  1.86s/it]
  3%|▎         | 398/14725 [13:10<7:21:59,  1.85s/it]
  3%|▎         | 399/14725 [13:11<7:18:40,  1.84s/it]
  3%|▎         | 400/14725 [13:13<7:24:43,  1.86s/it]
                                                     

  3%|▎         | 400/14725 [13:13<7:24:43,  1.86s/it]
  3%|▎         | 401/14725 [13:15<7:20:47,  1.85s/it]
  3%|▎         | 402/14725 [13:17<7:20:58,  1.85s/it]
  3%|▎         | 403/14725 [13:19<7:16:08,  1.83s/it]
  3%|▎         | 404/14725 [13:21<7:12:01,  1.81s/it]
  3%|▎         | 405/14725 [1
+0: {'loss': 0.948, 'grad_norm': 1.0821350158003564, 'learning_rate': 2e-05, 'epoch': 0.03}
+0: 3:22<7:14:11,  1.82s/it]
  3%|▎         | 406/14725 [13:24<7:15:51,  1.83s/it]
  3%|▎         | 407/14725 [13:26<7:19:21,  1.84s/it]
  3%|▎         | 408/14725 [13:28<7:15:43,  1.83s/it]
  3%|▎         | 409/14725 [13:30<7:11:26,  1.81s/it]
  3%|▎         | 410/14725 [13:32<7:11:22,  1.81s/it]
                                                     

  3%|▎         | 410/14725 [13:32<7:11:22,  1.81s/it]
  3%|▎         | 411/14725 [13:33<7:12:55,  1.81s/it]
  3%|▎         | 412/14725 [13:35<7:14:28,  1.82s/it]
  3%|▎         | 413/14725 [13:37<7:18:21,  1.84s/it]
  3%|▎         | 414/14725 [13:39<7:15:08,  1.82s/it]
  3%|▎         | 415/14725 [13:41<7:21:41,  1.85s/it]
  3%|▎         | 416/14725 [13:43<7:24:56,  1.87s/it]
  3%|▎         | 417/14725 [13:45<7:28:10,  1.88s/it]
  3%|▎         | 418/14725 [13:46<7:29:28,  1.89s/it]
  3%|▎         | 419/14725 [13:48<7:21:18,  1.85s/it]
  3%|▎         | 420/14725 [13:50<7:27:21,  1.88s/it]
                                                
+0: {'loss': 0.9691, 'grad_norm': 0.8887135075174087, 'learning_rate': 2e-05, 'epoch': 0.03}
+0: {'loss': 0.9746, 'grad_norm': 1.023340503907601, 'learning_rate': 2e-05, 'epoch': 0.03}
+0:      

  3%|▎         | 420/14725 [13:50<7:27:21,  1.88s/it]
  3%|▎         | 421/14725 [13:52<7:24:13,  1.86s/it]
  3%|▎         | 422/14725 [13:54<7:21:15,  1.85s/it]
  3%|▎         | 423/14725 [13:56<7:19:36,  1.84s/it]
  3%|▎         | 424/14725 [13:57<7:15:48,  1.83s/it]
  3%|▎         | 425/14725 [13:59<7:11:51,  1.81s/it]
  3%|▎         | 426/14725 [14:01<7:11:58,  1.81s/it]
  3%|▎         | 427/14725 [14:03<7:13:46,  1.82s/it]
  3%|▎         | 428/14725 [14:05<7:11:02,  1.81s/it]
  3%|▎         | 429/14725 [14:07<7:15:35,  1.83s/it]
  3%|▎         | 430/14725 [14:08<7:10:59,  1.81s/it]
                                                     

  3%|▎         | 430/14725 [14:08<7:10:59,  1.81s/it]
  3%|▎         | 431/14725 [14:10<7:10:38,  1.81s/it]
  3%|▎         | 432/14725 [14:12<7:15:06,  1.83s/it]
  3%|▎         | 433/14725 [14:14<7:15:31,  1.83s/it]
  3%|▎         | 434/14725 [14:16<7:23:29,  1.86s/it]
  3%|▎         | 435/14725 [14:18<7:22:02,  1.86s/it]
  3%|▎  
+0: {'loss': 0.9566, 'grad_norm': 1.1748495549250146, 'learning_rate': 2e-05, 'epoch': 0.03}
+0: {'loss': 0.9619, 'grad_norm': 1.061517175267452, 'learning_rate': 2e-05, 'epoch': 0.03}
+0:        | 436/14725 [14:19<7:16:04,  1.83s/it]
  3%|▎         | 437/14725 [14:21<7:17:00,  1.84s/it]
  3%|▎         | 438/14725 [14:23<7:19:42,  1.85s/it]
  3%|▎         | 439/14725 [14:25<7:17:51,  1.84s/it]
  3%|▎         | 440/14725 [14:27<7:20:21,  1.85s/it]
                                                     

  3%|▎         | 440/14725 [14:27<7:20:21,  1.85s/it]
  3%|▎         | 441/14725 [14:29<7:17:43,  1.84s/it]
  3%|▎         | 442/14725 [14:30<7:13:43,  1.82s/it]
  3%|▎         | 443/14725 [14:32<7:18:09,  1.84s/it]
  3%|▎         | 444/14725 [14:34<7:16:12,  1.83s/it]
  3%|▎         | 445/14725 [14:36<7:10:28,  1.81s/it]
  3%|▎         | 446/14725 [14:38<7:11:02,  1.81s/it]
  3%|▎         | 447/14725 [14:39<7:07:19,  1.80s/it]
  3%|▎         | 448/14725 [14:41<7:05:39,  1.79s/it]
  3%|▎         | 449/14725 [14:43<7:14:53,  1.83s/it]
  3%|▎         | 450/14725 [14:45<7:17:07,  1.84s/it]
                                                     

  3%|▎         | 450/14725
+0: {'loss': 0.9522, 'grad_norm': 0.9097775323292734, 'learning_rate': 2e-05, 'epoch': 0.03}
+0:  [14:45<7:17:07,  1.84s/it]
  3%|▎         | 451/14725 [14:47<7:14:15,  1.83s/it]
  3%|▎         | 452/14725 [14:49<7:14:07,  1.82s/it]
  3%|▎         | 453/14725 [14:50<7:09:49,  1.81s/it]
  3%|▎         | 454/14725 [14:52<7:11:53,  1.82s/it]
  3%|▎         | 455/14725 [14:54<7:16:13,  1.83s/it]
  3%|▎         | 456/14725 [14:56<7:16:17,  1.83s/it]
  3%|▎         | 457/14725 [14:58<7:19:11,  1.85s/it]
  3%|▎         | 458/14725 [15:00<7:15:08,  1.83s/it]
  3%|▎         | 459/14725 [15:01<7:11:35,  1.82s/it]
  3%|▎         | 460/14725 [15:03<7:10:49,  1.81s/it]
                                                     

  3%|▎         | 460/14725 [15:03<7:10:49,  1.81s/it]
  3%|▎         | 461/14725 [15:05<7:08:41,  1.80s/it]
  3%|▎         | 462/14725 [15:07<7:13:10,  1.82s/it]
  3%|▎         | 463/14725 [15:09<7:14:12,  1.83s/it]
  3%|▎         | 464/14725 [15:10<7:14:33,  1.83s/it]
  3%|▎         | 465/14725 [15:12<7:22:34,  1.86s/it]
  3%|▎         | 466/14725 [15:14<7:17:45, 
+0: {'loss': 0.9747, 'grad_norm': 0.877833632184112, 'learning_rate': 2e-05, 'epoch': 0.03}
+0: {'loss': 0.9506, 'grad_norm': 0.9234233840009626, 'learning_rate': 2e-05, 'epoch': 0.03}
+0:  1.84s/it]
  3%|▎         | 467/14725 [15:16<7:14:34,  1.83s/it]
  3%|▎         | 468/14725 [15:18<7:15:15,  1.83s/it]
  3%|▎         | 469/14725 [15:20<7:12:27,  1.82s/it]
  3%|▎         | 470/14725 [15:22<7:20:30,  1.85s/it]
                                                     

  3%|▎         | 470/14725 [15:22<7:20:30,  1.85s/it]
  3%|▎         | 471/14725 [15:23<7:21:08,  1.86s/it]
  3%|▎         | 472/14725 [15:25<7:18:22,  1.85s/it]
  3%|▎         | 473/14725 [15:27<7:18:45,  1.85s/it]
  3%|▎         | 474/14725 [15:29<7:15:52,  1.84s/it]
  3%|▎         | 475/14725 [15:31<7:13:05,  1.82s/it]
  3%|▎         | 476/14725 [15:33<7:12:57,  1.82s/it]
  3%|▎         | 477/14725 [15:34<7:09:01,  1.81s/it]
  3%|▎         | 478/14725 [15:36<7:11:39,  1.82s/it]
  3%|▎         | 479/14725 [15:38<7:11:14,  1.82s/it]
  3%|▎         | 480/14725 [15:40<7:10:16,  1.81s/it]
                                                     

  3%|▎         | 480/14725 [15:40<7:10:16,  1.81s/it]
  3%|�
+0: {'loss': 0.9429, 'grad_norm': 1.063487398605876, 'learning_rate': 2e-05, 'epoch': 0.03}
+0: �         | 481/14725 [15:42<7:10:24,  1.81s/it]
  3%|▎         | 482/14725 [15:43<7:09:50,  1.81s/it]
  3%|▎         | 483/14725 [15:45<7:10:20,  1.81s/it]
  3%|▎         | 484/14725 [15:47<7:10:51,  1.82s/it]
  3%|▎         | 485/14725 [15:49<7:12:53,  1.82s/it]
  3%|▎         | 486/14725 [15:51<7:13:28,  1.83s/it]
  3%|▎         | 487/14725 [15:53<7:13:22,  1.83s/it]
  3%|▎         | 488/14725 [15:54<7:12:26,  1.82s/it]
  3%|▎         | 489/14725 [15:56<7:15:26,  1.84s/it]
  3%|▎         | 490/14725 [15:58<7:13:58,  1.83s/it]
                                                     

  3%|▎         | 490/14725 [15:58<7:13:58,  1.83s/it]
  3%|▎         | 491/14725 [16:00<7:15:56,  1.84s/it]
  3%|▎         | 492/14725 [16:02<7:10:43,  1.82s/it]
  3%|▎         | 493/14725 [16:03<7:08:47,  1.81s/it]
  3%|▎         | 494/14725 [16:05<7:12:18,  1.82s/it]
  3%|▎         | 495/14725 [16:07<7:14:31,  1.83s/it]
  3%|▎         | 496/14725 [16:09<7:23:37,  1.87s/it]
  3%|▎         | 497/1
+0: {'loss': 0.959, 'grad_norm': 0.8935243508590848, 'learning_rate': 2e-05, 'epoch': 0.03}
+0: {'loss': 0.9729, 'grad_norm': 1.0562396724466054, 'learning_rate': 2e-05, 'epoch': 0.03}
+0: 4725 [16:11<7:17:06,  1.84s/it]
  3%|▎         | 498/14725 [16:13<7:15:38,  1.84s/it]
  3%|▎         | 499/14725 [16:15<7:15:19,  1.84s/it]
  3%|▎         | 500/14725 [16:16<7:17:02,  1.84s/it]
                                                     

  3%|▎         | 500/14725 [16:16<7:17:02,  1.84s/it]
  3%|▎         | 501/14725 [16:18<7:14:47,  1.83s/it]
  3%|▎         | 502/14725 [16:20<7:18:07,  1.85s/it]
  3%|▎         | 503/14725 [16:22<7:16:30,  1.84s/it]
  3%|▎         | 504/14725 [16:24<7:10:57,  1.82s/it]
  3%|▎         | 505/14725 [16:26<7:16:49,  1.84s/it]
  3%|▎         | 506/14725 [16:27<7:14:55,  1.84s/it]
  3%|▎         | 507/14725 [16:29<7:17:01,  1.84s/it]
  3%|▎         | 508/14725 [16:31<7:14:42,  1.83s/it]
  3%|▎         | 509/14725 [16:33<7:16:15,  1.84s/it]
  3%|▎         | 510/14725 [16:35<7:17:03,  1.84s/it]
                                                     

  3%|▎         | 510/14725 [16:35<7:17:03,  1.84s/it]
  3%|▎         | 511/14725 [16:37<7:15:4
+0: {'loss': 0.9576, 'grad_norm': 1.0402616595402672, 'learning_rate': 2e-05, 'epoch': 0.04}
+0: 0,  1.84s/it]
  3%|▎         | 512/14725 [16:38<7:13:15,  1.83s/it]
  3%|▎         | 513/14725 [16:40<7:14:23,  1.83s/it]
  3%|▎         | 514/14725 [16:42<7:12:23,  1.83s/it]
  3%|▎         | 515/14725 [16:44<7:13:13,  1.83s/it]
  4%|▎         | 516/14725 [16:46<7:14:38,  1.84s/it]
  4%|▎         | 517/14725 [16:48<7:26:39,  1.89s/it]
  4%|▎         | 518/14725 [16:50<7:26:42,  1.89s/it]
  4%|▎         | 519/14725 [16:51<7:20:03,  1.86s/it]
  4%|▎         | 520/14725 [16:53<7:14:28,  1.84s/it]
                                                     

  4%|▎         | 520/14725 [16:53<7:14:28,  1.84s/it]
  4%|▎         | 521/14725 [16:55<7:10:17,  1.82s/it]
  4%|▎         | 522/14725 [16:57<7:11:22,  1.82s/it]
  4%|▎         | 523/14725 [16:59<7:09:43,  1.82s/it]
  4%|▎         | 524/14725 [17:01<7:20:28,  1.86s/it]
  4%|▎         | 525/14725 [17:02<7:16:27,  1.84s/it]
  4%|▎         | 526/14725 [17:04<7:15:21,  1.84s/it]
  4%|▎         | 527/14725 [17:06<7:14:46,  1.84s/it]
  4
+0: {'loss': 0.9551, 'grad_norm': 0.9712051817457336, 'learning_rate': 2e-05, 'epoch': 0.04}
+0: {'loss': 0.9559, 'grad_norm': 0.9381529664634524, 'learning_rate': 2e-05, 'epoch': 0.04}
+0: %|▎         | 528/14725 [17:08<7:14:28,  1.84s/it]
  4%|▎         | 529/14725 [17:10<7:19:20,  1.86s/it]
  4%|▎         | 530/14725 [17:12<7:15:35,  1.84s/it]
                                                     

  4%|▎         | 530/14725 [17:12<7:15:35,  1.84s/it]
  4%|▎         | 531/14725 [17:13<7:13:09,  1.83s/it]
  4%|▎         | 532/14725 [17:15<7:12:33,  1.83s/it]
  4%|▎         | 533/14725 [17:17<7:15:26,  1.84s/it]
  4%|▎         | 534/14725 [17:19<7:13:54,  1.83s/it]
  4%|▎         | 535/14725 [17:21<7:14:28,  1.84s/it]
  4%|▎         | 536/14725 [17:23<7:14:24,  1.84s/it]
  4%|▎         | 537/14725 [17:24<7:11:15,  1.82s/it]
  4%|▎         | 538/14725 [17:26<7:11:14,  1.82s/it]
  4%|▎         | 539/14725 [17:28<7:12:35,  1.83s/it]
  4%|▎         | 540/14725 [17:30<7:12:58,  1.83s/it]
                                                     

  4%|▎         | 540/14725 [17:30<7:12:58,  1.83s/it]
  4%|▎         | 541/14725 [17:32<7:16:47,  1.85s/it]
  4%|▎         | 54
+0: {'loss': 0.9559, 'grad_norm': 0.9063902093797513, 'learning_rate': 2e-05, 'epoch': 0.04}
+0: 2/14725 [17:34<7:19:51,  1.86s/it]
  4%|▎         | 543/14725 [17:36<7:18:15,  1.85s/it]
  4%|▎         | 544/14725 [17:37<7:19:43,  1.86s/it]
  4%|▎         | 545/14725 [17:39<7:15:20,  1.84s/it]
  4%|▎         | 546/14725 [17:41<7:12:51,  1.83s/it]
  4%|▎         | 547/14725 [17:43<7:13:43,  1.84s/it]
  4%|▎         | 548/14725 [17:45<7:13:54,  1.84s/it]
  4%|▎         | 549/14725 [17:46<7:12:26,  1.83s/it]
  4%|▎         | 550/14725 [17:48<7:12:25,  1.83s/it]
                                                     

  4%|▎         | 550/14725 [17:48<7:12:25,  1.83s/it]
  4%|▎         | 551/14725 [17:50<7:16:46,  1.85s/it]
  4%|▎         | 552/14725 [17:52<7:29:52,  1.90s/it]
  4%|▍         | 553/14725 [17:54<7:27:28,  1.89s/it]
  4%|▍         | 554/14725 [17:56<7:21:45,  1.87s/it]
  4%|▍         | 555/14725 [17:58<7:27:36,  1.90s/it]
  4%|▍         | 556/14725 [18:00<7:24:43,  1.88s/it]
  4%|▍         | 557/14725 [18:02<7:17:13,  1.85s/it]
  4%|▍         | 558/14725 [18:03<7:
+0: {'loss': 0.9475, 'grad_norm': 0.9603848035660635, 'learning_rate': 2e-05, 'epoch': 0.04}
+0: {'loss': 0.9457, 'grad_norm': 0.94632125960494, 'learning_rate': 2e-05, 'epoch': 0.04}
+0: 14:35,  1.84s/it]
  4%|▍         | 559/14725 [18:05<7:13:20,  1.84s/it]
  4%|▍         | 560/14725 [18:07<7:08:39,  1.82s/it]
                                                     

  4%|▍         | 560/14725 [18:07<7:08:39,  1.82s/it]
  4%|▍         | 561/14725 [18:09<7:09:49,  1.82s/it]
  4%|▍         | 562/14725 [18:11<7:06:00,  1.80s/it]
  4%|▍         | 563/14725 [18:12<7:06:27,  1.81s/it]
  4%|▍         | 564/14725 [18:14<7:07:05,  1.81s/it]
  4%|▍         | 565/14725 [18:16<7:05:26,  1.80s/it]
  4%|▍         | 566/14725 [18:18<7:09:18,  1.82s/it]
  4%|▍         | 567/14725 [18:20<7:12:07,  1.83s/it]
  4%|▍         | 568/14725 [18:22<7:18:14,  1.86s/it]
  4%|▍         | 569/14725 [18:24<7:28:05,  1.90s/it]
  4%|▍         | 570/14725 [18:25<7:25:14,  1.89s/it]
                                                     

  4%|▍         | 570/14725 [18:25<7:25:14,  1.89s/it]
  4%|▍         | 571/14725 [18:27<7:23:08,  1.88s/it]
  4%|▍         | 572/14725 [18:29<7:25:16,  1.89s/it]
+0: {'loss': 0.9417, 'grad_norm': 0.9905278614546392, 'learning_rate': 2e-05, 'epoch': 0.04}
+0:   4%|▍         | 573/14725 [18:31<7:18:12,  1.86s/it]
  4%|▍         | 574/14725 [18:33<7:20:29,  1.87s/it]
  4%|▍         | 575/14725 [18:35<7:23:59,  1.88s/it]
  4%|▍         | 576/14725 [18:37<7:17:33,  1.86s/it]
  4%|▍         | 577/14725 [18:39<7:22:32,  1.88s/it]
  4%|▍         | 578/14725 [18:41<7:34:25,  1.93s/it]
  4%|▍         | 579/14725 [18:42<7:26:12,  1.89s/it]
  4%|▍         | 580/14725 [18:44<7:29:36,  1.91s/it]
                                                     

  4%|▍         | 580/14725 [18:44<7:29:36,  1.91s/it]
  4%|▍         | 581/14725 [18:46<7:26:19,  1.89s/it]
  4%|▍         | 582/14725 [18:48<7:26:02,  1.89s/it]
  4%|▍         | 583/14725 [18:50<7:22:26,  1.88s/it]
  4%|▍         | 584/14725 [18:52<7:19:05,  1.86s/it]
  4%|▍         | 585/14725 [18:54<7:24:24,  1.89s/it]
  4%|▍         | 586/14725 [18:56<7:31:48,  1.92s/it]
  4%|▍         | 587/14725 [18:58<7:37:34,  1.94s/it]
  4%|▍         | 588/14725 [19:00<7:34:14,  1.93s/it]
  4%|▍         
+0: {'loss': 0.961, 'grad_norm': 0.9314024971600879, 'learning_rate': 2e-05, 'epoch': 0.04}
+0: {'loss': 0.9398, 'grad_norm': 0.9771363859902632, 'learning_rate': 2e-05, 'epoch': 0.04}
+0: | 589/14725 [19:01<7:27:27,  1.90s/it]
  4%|▍         | 590/14725 [19:03<7:30:49,  1.91s/it]
                                                     

  4%|▍         | 590/14725 [19:03<7:30:49,  1.91s/it]
  4%|▍         | 591/14725 [19:05<7:27:20,  1.90s/it]
  4%|▍         | 592/14725 [19:07<7:24:22,  1.89s/it]
  4%|▍         | 593/14725 [19:09<7:22:32,  1.88s/it]
  4%|▍         | 594/14725 [19:11<7:15:00,  1.85s/it]
  4%|▍         | 595/14725 [19:13<7:15:36,  1.85s/it]
  4%|▍         | 596/14725 [19:14<7:13:40,  1.84s/it]
  4%|▍         | 597/14725 [19:16<7:14:27,  1.85s/it]
  4%|▍         | 598/14725 [19:18<7:16:10,  1.85s/it]
  4%|▍         | 599/14725 [19:20<7:14:54,  1.85s/it]
  4%|▍         | 600/14725 [19:22<7:10:47,  1.83s/it]
                                                     

  4%|▍         | 600/14725 [19:22<7:10:47,  1.83s/it]
  4%|▍         | 601/14725 [19:24<7:25:20,  1.89s/it]
  4%|▍         | 602/14725 [19:26<7:23:17,  1.88s/it]
  4%|▍         | 603/14725 [19:27
+0: {'loss': 0.9489, 'grad_norm': 0.939744816048822, 'learning_rate': 2e-05, 'epoch': 0.04}
+0: <7:19:56,  1.87s/it]
  4%|▍         | 604/14725 [19:29<7:18:58,  1.87s/it]
  4%|▍         | 605/14725 [19:31<7:20:42,  1.87s/it]
  4%|▍         | 606/14725 [19:33<7:23:01,  1.88s/it]
  4%|▍         | 607/14725 [19:35<7:27:25,  1.90s/it]
  4%|▍         | 608/14725 [19:37<7:26:12,  1.90s/it]
  4%|▍         | 609/14725 [19:39<7:25:17,  1.89s/it]
  4%|▍         | 610/14725 [19:41<7:25:09,  1.89s/it]
                                                     

  4%|▍         | 610/14725 [19:41<7:25:09,  1.89s/it]
  4%|▍         | 611/14725 [19:43<7:17:55,  1.86s/it]
  4%|▍         | 612/14725 [19:44<7:19:32,  1.87s/it]
  4%|▍         | 613/14725 [19:46<7:22:32,  1.88s/it]
  4%|▍         | 614/14725 [19:48<7:20:32,  1.87s/it]
  4%|▍         | 615/14725 [19:50<7:14:51,  1.85s/it]
  4%|▍         | 616/14725 [19:52<7:20:19,  1.87s/it]
  4%|▍         | 617/14725 [19:54<7:18:06,  1.86s/it]
  4%|▍         | 618/14725 [19:56<7:17:38,  1.86s/it]
  4%|▍         | 619/14725 [19:57<7:19:22,  1.87s/
+0: {'loss': 0.9477, 'grad_norm': 1.0219003980162547, 'learning_rate': 2e-05, 'epoch': 0.04}
+0: {'loss': 0.9456, 'grad_norm': 1.1107455201073964, 'learning_rate': 2e-05, 'epoch': 0.04}
+0: it]
  4%|▍         | 620/14725 [19:59<7:13:57,  1.85s/it]
                                                     

  4%|▍         | 620/14725 [19:59<7:13:57,  1.85s/it]
  4%|▍         | 621/14725 [20:01<7:14:37,  1.85s/it]
  4%|▍         | 622/14725 [20:03<7:30:24,  1.92s/it]
  4%|▍         | 623/14725 [20:05<7:25:37,  1.90s/it]
  4%|▍         | 624/14725 [20:07<7:22:49,  1.88s/it]
  4%|▍         | 625/14725 [20:09<7:25:20,  1.90s/it]
  4%|▍         | 626/14725 [20:11<7:20:02,  1.87s/it]
  4%|▍         | 627/14725 [20:13<7:21:01,  1.88s/it]
  4%|▍         | 628/14725 [20:14<7:20:03,  1.87s/it]
  4%|▍         | 629/14725 [20:16<7:14:18,  1.85s/it]
  4%|▍         | 630/14725 [20:18<7:17:09,  1.86s/it]
                                                     

  4%|▍         | 630/14725 [20:18<7:17:09,  1.86s/it]
  4%|▍         | 631/14725 [20:20<7:14:19,  1.85s/it]
  4%|▍         | 632/14725 [20:22<7:14:25,  1.85s/it]
  4%|▍         | 633/14725 [20:24<7:24:10,  1.89s/it]
  4%|▍      
+0: {'loss': 0.9471, 'grad_norm': 0.8636253905215894, 'learning_rate': 2e-05, 'epoch': 0.04}
+0:    | 634/14725 [20:26<7:19:00,  1.87s/it]
  4%|▍         | 635/14725 [20:27<7:16:15,  1.86s/it]
  4%|▍         | 636/14725 [20:29<7:16:33,  1.86s/it]
  4%|▍         | 637/14725 [20:31<7:19:34,  1.87s/it]
  4%|▍         | 638/14725 [20:33<7:33:15,  1.93s/it]
  4%|▍         | 639/14725 [20:35<7:30:57,  1.92s/it]
  4%|▍         | 640/14725 [20:37<7:25:21,  1.90s/it]
                                                     

  4%|▍         | 640/14725 [20:37<7:25:21,  1.90s/it]
  4%|▍         | 641/14725 [20:39<7:22:03,  1.88s/it]
  4%|▍         | 642/14725 [20:41<7:23:45,  1.89s/it]
  4%|▍         | 643/14725 [20:43<7:33:29,  1.93s/it]
  4%|▍         | 644/14725 [20:45<7:29:04,  1.91s/it]
  4%|▍         | 645/14725 [20:47<7:28:27,  1.91s/it]
  4%|▍         | 646/14725 [20:48<7:23:41,  1.89s/it]
  4%|▍         | 647/14725 [20:50<7:22:28,  1.89s/it]
  4%|▍         | 648/14725 [20:52<7:33:56,  1.93s/it]
  4%|▍         | 649/14725 [20:54<7:37:11,  1.95s/it]
  4%|▍         | 650/14725 [2
+0: {'loss': 0.9439, 'grad_norm': 0.9643281740484851, 'learning_rate': 2e-05, 'epoch': 0.04}
+0: {'loss': 0.9328, 'grad_norm': 1.0307559050560773, 'learning_rate': 2e-05, 'epoch': 0.04}
+0: 0:56<7:31:10,  1.92s/it]
                                                     

  4%|▍         | 650/14725 [20:56<7:31:10,  1.92s/it]
  4%|▍         | 651/14725 [20:58<7:29:34,  1.92s/it]
  4%|▍         | 652/14725 [21:00<7:20:46,  1.88s/it]
  4%|▍         | 653/14725 [21:02<7:22:29,  1.89s/it]
  4%|▍         | 654/14725 [21:04<7:25:26,  1.90s/it]
  4%|▍         | 655/14725 [21:06<7:22:37,  1.89s/it]
  4%|▍         | 656/14725 [21:07<7:20:23,  1.88s/it]
  4%|▍         | 657/14725 [21:09<7:25:00,  1.90s/it]
  4%|▍         | 658/14725 [21:11<7:20:39,  1.88s/it]
  4%|▍         | 659/14725 [21:13<7:19:01,  1.87s/it]
  4%|▍         | 660/14725 [21:15<7:18:52,  1.87s/it]
                                                     

  4%|▍         | 660/14725 [21:15<7:18:52,  1.87s/it]
  4%|▍         | 661/14725 [21:17<7:16:54,  1.86s/it]
  4%|▍         | 662/14725 [21:19<7:15:27,  1.86s/it]
  5%|▍         | 663/14725 [21:21<7:21:10,  1.88s/it]
  5%|▍         | 664/14725 [21:22<7:15:08,  1.8
+0: {'loss': 0.9232, 'grad_norm': 0.945344491018314, 'learning_rate': 2e-05, 'epoch': 0.05}
+0: 6s/it]
  5%|▍         | 665/14725 [21:24<7:15:58,  1.86s/it]
  5%|▍         | 666/14725 [21:26<7:16:47,  1.86s/it]
  5%|▍         | 667/14725 [21:28<7:22:27,  1.89s/it]
  5%|▍         | 668/14725 [21:30<7:17:15,  1.87s/it]
  5%|▍         | 669/14725 [21:32<7:14:46,  1.86s/it]
  5%|▍         | 670/14725 [21:33<7:15:38,  1.86s/it]
                                                     

  5%|▍         | 670/14725 [21:34<7:15:38,  1.86s/it]
  5%|▍         | 671/14725 [21:35<7:15:31,  1.86s/it]
  5%|▍         | 672/14725 [21:37<7:16:57,  1.87s/it]
  5%|▍         | 673/14725 [21:39<7:20:30,  1.88s/it]
  5%|▍         | 674/14725 [21:41<7:19:54,  1.88s/it]
  5%|▍         | 675/14725 [21:43<7:18:19,  1.87s/it]
  5%|▍         | 676/14725 [21:45<7:21:14,  1.88s/it]
  5%|▍         | 677/14725 [21:47<7:20:43,  1.88s/it]
  5%|▍         | 678/14725 [21:49<7:19:45,  1.88s/it]
  5%|▍         | 679/14725 [21:50<7:16:50,  1.87s/it]
  5%|▍         | 680/14725 [21:52<7:14:49,  1.86s/it]
          
+0: {'loss': 0.9453, 'grad_norm': 0.995191070760294, 'learning_rate': 2e-05, 'epoch': 0.05}
+0: {'loss': 0.9403, 'grad_norm': 1.005483663789468, 'learning_rate': 2e-05, 'epoch': 0.05}
+0:                                            

  5%|▍         | 680/14725 [21:52<7:14:49,  1.86s/it]
  5%|▍         | 681/14725 [21:54<7:19:10,  1.88s/it]
  5%|▍         | 682/14725 [21:56<7:16:31,  1.87s/it]
  5%|▍         | 683/14725 [21:58<7:16:02,  1.86s/it]
  5%|▍         | 684/14725 [22:00<7:21:45,  1.89s/it]
  5%|▍         | 685/14725 [22:02<7:29:12,  1.92s/it]
  5%|▍         | 686/14725 [22:04<7:24:52,  1.90s/it]
  5%|▍         | 687/14725 [22:05<7:20:41,  1.88s/it]
  5%|▍         | 688/14725 [22:07<7:24:26,  1.90s/it]
  5%|▍         | 689/14725 [22:09<7:24:38,  1.90s/it]
  5%|▍         | 690/14725 [22:11<7:19:54,  1.88s/it]
                                                     

  5%|▍         | 690/14725 [22:11<7:19:54,  1.88s/it]
  5%|▍         | 691/14725 [22:13<7:17:44,  1.87s/it]
  5%|▍         | 692/14725 [22:15<7:22:31,  1.89s/it]
  5%|▍         | 693/14725 [22:17<7:19:36,  1.88s/it]
  5%|▍         | 694/14725 [22:19<7:17:56,  1.87s/it]
  5%|▍         | 695/14725
+0: {'loss': 0.9286, 'grad_norm': 0.8352318000665129, 'learning_rate': 2e-05, 'epoch': 0.05}
+0:  [22:21<7:18:08,  1.87s/it]
  5%|▍         | 696/14725 [22:22<7:15:35,  1.86s/it]
  5%|▍         | 697/14725 [22:24<7:16:16,  1.87s/it]
  5%|▍         | 698/14725 [22:26<7:27:09,  1.91s/it]
  5%|▍         | 699/14725 [22:28<7:26:48,  1.91s/it]
  5%|▍         | 700/14725 [22:30<7:24:47,  1.90s/it]
                                                     

  5%|▍         | 700/14725 [22:30<7:24:47,  1.90s/it]
  5%|▍         | 701/14725 [22:32<7:19:36,  1.88s/it]
  5%|▍         | 702/14725 [22:34<7:17:40,  1.87s/it]
  5%|▍         | 703/14725 [22:36<7:17:34,  1.87s/it]
  5%|▍         | 704/14725 [22:37<7:13:23,  1.85s/it]
  5%|▍         | 705/14725 [22:39<7:18:50,  1.88s/it]
  5%|▍         | 706/14725 [22:41<7:26:09,  1.91s/it]
  5%|▍         | 707/14725 [22:43<7:33:42,  1.94s/it]
  5%|▍         | 708/14725 [22:45<7:34:32,  1.95s/it]
  5%|▍         | 709/14725 [22:47<7:24:15,  1.90s/it]
  5%|▍         | 710/14725 [22:49<7:22:10,  1.89s/it]
                                             
+0: {'loss': 0.9593, 'grad_norm': 0.857234495059117, 'learning_rate': 2e-05, 'epoch': 0.05}
+0: {'loss': 0.9428, 'grad_norm': 1.1429633295718697, 'learning_rate': 2e-05, 'epoch': 0.05}
+0:         

  5%|▍         | 710/14725 [22:49<7:22:10,  1.89s/it]
  5%|▍         | 711/14725 [22:51<7:16:03,  1.87s/it]
  5%|▍         | 712/14725 [22:53<7:14:44,  1.86s/it]
  5%|▍         | 713/14725 [22:55<7:15:27,  1.86s/it]
  5%|▍         | 714/14725 [22:56<7:18:33,  1.88s/it]
  5%|▍         | 715/14725 [22:58<7:17:14,  1.87s/it]
  5%|▍         | 716/14725 [23:00<7:17:35,  1.87s/it]
  5%|▍         | 717/14725 [23:02<7:14:55,  1.86s/it]
  5%|▍         | 718/14725 [23:04<7:38:21,  1.96s/it]
  5%|▍         | 719/14725 [23:06<7:32:09,  1.94s/it]
  5%|▍         | 720/14725 [23:08<7:29:38,  1.93s/it]
                                                     

  5%|▍         | 720/14725 [23:08<7:29:38,  1.93s/it]
  5%|▍         | 721/14725 [23:10<7:22:38,  1.90s/it]
  5%|▍         | 722/14725 [23:12<7:33:34,  1.94s/it]
  5%|▍         | 723/14725 [23:14<7:57:48,  2.05s/it]
  5%|▍         | 724/14725 [23:16<7:57:44,  2.05s/it]
  5%|▍         | 725/14725 [23:18<7:42:18,  1.98s/it]
  5%|�
+0: {'loss': 0.9397, 'grad_norm': 0.8578553196914328, 'learning_rate': 2e-05, 'epoch': 0.05}
+0: {'loss': 0.947, 'grad_norm': 0.965581796037372, 'learning_rate': 2e-05, 'epoch': 0.05}
+0: �         | 726/14725 [23:20<7:35:48,  1.95s/it]
  5%|▍         | 727/14725 [23:22<7:37:26,  1.96s/it]
  5%|▍         | 728/14725 [23:24<7:39:48,  1.97s/it]
  5%|▍         | 729/14725 [23:26<7:34:46,  1.95s/it]
  5%|▍         | 730/14725 [23:28<7:29:55,  1.93s/it]
                                                     

  5%|▍         | 730/14725 [23:28<7:29:55,  1.93s/it]
  5%|▍         | 731/14725 [23:30<7:28:01,  1.92s/it]
  5%|▍         | 732/14725 [23:31<7:24:46,  1.91s/it]
  5%|▍         | 733/14725 [23:33<7:25:40,  1.91s/it]
  5%|▍         | 734/14725 [23:35<7:27:34,  1.92s/it]
  5%|▍         | 735/14725 [23:37<7:24:41,  1.91s/it]
  5%|▍         | 736/14725 [23:39<7:26:22,  1.91s/it]
  5%|▌         | 737/14725 [23:41<7:29:31,  1.93s/it]
  5%|▌         | 738/14725 [23:43<7:26:33,  1.92s/it]
  5%|▌         | 739/14725 [23:45<7:47:40,  2.01s/it]
  5%|▌         | 740/14725 [23:47<7:38:11,  1.97s/it]
                                                     

  5%|▌         | 740/14
+0: {'loss': 0.9238, 'grad_norm': 0.8988611675161893, 'learning_rate': 2e-05, 'epoch': 0.05}
+0: 725 [23:47<7:38:11,  1.97s/it]
  5%|▌         | 741/14725 [23:49<7:45:10,  2.00s/it]
  5%|▌         | 742/14725 [23:51<7:42:28,  1.98s/it]
  5%|▌         | 743/14725 [23:53<7:35:08,  1.95s/it]
  5%|▌         | 744/14725 [23:55<7:41:17,  1.98s/it]
  5%|▌         | 745/14725 [23:57<7:41:47,  1.98s/it]
  5%|▌         | 746/14725 [23:59<7:37:59,  1.97s/it]
  5%|▌         | 747/14725 [24:01<7:32:11,  1.94s/it]
  5%|▌         | 748/14725 [24:03<7:40:33,  1.98s/it]
  5%|▌         | 749/14725 [24:05<7:38:26,  1.97s/it]
  5%|▌         | 750/14725 [24:07<7:27:15,  1.92s/it]
                                                     

  5%|▌         | 750/14725 [24:07<7:27:15,  1.92s/it]
  5%|▌         | 751/14725 [24:09<7:42:42,  1.99s/it]
  5%|▌         | 752/14725 [24:11<7:35:47,  1.96s/it]
  5%|▌         | 753/14725 [24:12<7:30:06,  1.93s/it]
  5%|▌         | 754/14725 [24:14<7:27:48,  1.92s/it]
  5%|▌         | 755/14725 [24:16<7:23:47,  1.91s/it]
  5%|▌         | 756/14725 [24:18<7:21:2
+0: {'loss': 0.9437, 'grad_norm': 0.8997065885288374, 'learning_rate': 2e-05, 'epoch': 0.05}
+0: {'loss': 0.9442, 'grad_norm': 0.846692091117542, 'learning_rate': 2e-05, 'epoch': 0.05}
+0: 5,  1.90s/it]
  5%|▌         | 757/14725 [24:20<7:25:01,  1.91s/it]
  5%|▌         | 758/14725 [24:22<7:22:49,  1.90s/it]
  5%|▌         | 759/14725 [24:24<7:17:16,  1.88s/it]
  5%|▌         | 760/14725 [24:26<7:15:59,  1.87s/it]
                                                     

  5%|▌         | 760/14725 [24:26<7:15:59,  1.87s/it]
  5%|▌         | 761/14725 [24:28<7:18:04,  1.88s/it]
  5%|▌         | 762/14725 [24:29<7:21:03,  1.90s/it]
  5%|▌         | 763/14725 [24:31<7:16:40,  1.88s/it]
  5%|▌         | 764/14725 [24:33<7:12:38,  1.86s/it]
  5%|▌         | 765/14725 [24:35<7:17:13,  1.88s/it]
  5%|▌         | 766/14725 [24:37<7:21:27,  1.90s/it]
  5%|▌         | 767/14725 [24:39<7:27:20,  1.92s/it]
  5%|▌         | 768/14725 [24:41<7:22:15,  1.90s/it]
  5%|▌         | 769/14725 [24:43<7:18:47,  1.89s/it]
  5%|▌         | 770/14725 [24:45<7:17:15,  1.88s/it]
                                                     

  5%|▌         | 770/14725 [24:45<7:17:15,  1.88s/it]
  5%
+0: {'loss': 0.923, 'grad_norm': 1.0559648566076567, 'learning_rate': 2e-05, 'epoch': 0.05}
+0: |▌         | 771/14725 [24:46<7:17:55,  1.88s/it]
  5%|▌         | 772/14725 [24:48<7:19:04,  1.89s/it]
  5%|▌         | 773/14725 [24:50<7:22:25,  1.90s/it]
  5%|▌         | 774/14725 [24:52<7:31:39,  1.94s/it]
  5%|▌         | 775/14725 [24:54<7:25:45,  1.92s/it]
  5%|▌         | 776/14725 [24:56<7:21:30,  1.90s/it]
  5%|▌         | 777/14725 [24:58<7:20:41,  1.90s/it]
  5%|▌         | 778/14725 [25:00<7:17:30,  1.88s/it]
  5%|▌         | 779/14725 [25:02<7:15:45,  1.87s/it]
  5%|▌         | 780/14725 [25:03<7:14:42,  1.87s/it]
                                                     

  5%|▌         | 780/14725 [25:03<7:14:42,  1.87s/it]
  5%|▌         | 781/14725 [25:05<7:16:38,  1.88s/it]
  5%|▌         | 782/14725 [25:07<7:19:50,  1.89s/it]
  5%|▌         | 783/14725 [25:09<7:20:09,  1.89s/it]
  5%|▌         | 784/14725 [25:11<7:38:53,  1.98s/it]
  5%|▌         | 785/14725 [25:13<7:36:49,  1.97s/it]
  5%|▌         | 786/14725 [25:15<7:29:35,  1.94s/it]
  5%|▌         | 78
+0: {'loss': 0.915, 'grad_norm': 0.885283182589103, 'learning_rate': 2e-05, 'epoch': 0.05}
+0: {'loss': 0.9398, 'grad_norm': 0.9773685787362743, 'learning_rate': 2e-05, 'epoch': 0.05}
+0: 7/14725 [25:17<7:27:02,  1.92s/it]
  5%|▌         | 788/14725 [25:19<7:23:11,  1.91s/it]
  5%|▌         | 789/14725 [25:21<7:25:33,  1.92s/it]
  5%|▌         | 790/14725 [25:23<7:20:56,  1.90s/it]
                                                     

  5%|▌         | 790/14725 [25:23<7:20:56,  1.90s/it]
  5%|▌         | 791/14725 [25:25<7:18:01,  1.89s/it]
  5%|▌         | 792/14725 [25:27<7:34:00,  1.96s/it]
  5%|▌         | 793/14725 [25:29<7:27:48,  1.93s/it]
  5%|▌         | 794/14725 [25:30<7:23:41,  1.91s/it]
  5%|▌         | 795/14725 [25:32<7:21:01,  1.90s/it]
  5%|▌         | 796/14725 [25:34<7:18:26,  1.89s/it]
  5%|▌         | 797/14725 [25:36<7:14:27,  1.87s/it]
  5%|▌         | 798/14725 [25:38<7:15:13,  1.88s/it]
  5%|▌         | 799/14725 [25:40<7:16:36,  1.88s/it]
  5%|▌         | 800/14725 [25:42<7:14:57,  1.87s/it]
                                                     

  5%|▌         | 800/14725 [25:42<7:14:57,  1.87s/it]
  5%|▌         | 801/14725 [25:44<7:1
+0: {'loss': 0.9227, 'grad_norm': 0.9529369726190446, 'learning_rate': 2e-05, 'epoch': 0.06}
+0: 5:32,  1.88s/it]
  5%|▌         | 802/14725 [25:45<7:16:47,  1.88s/it]
  5%|▌         | 803/14725 [25:47<7:18:44,  1.89s/it]
  5%|▌         | 804/14725 [25:49<7:15:13,  1.88s/it]
  5%|▌         | 805/14725 [25:51<7:18:35,  1.89s/it]
  5%|▌         | 806/14725 [25:53<7:20:21,  1.90s/it]
  5%|▌         | 807/14725 [25:55<7:26:48,  1.93s/it]
  5%|▌         | 808/14725 [25:57<7:36:15,  1.97s/it]
  5%|▌         | 809/14725 [25:59<8:06:28,  2.10s/it]
  6%|▌         | 810/14725 [26:01<7:55:05,  2.05s/it]
                                                     

  6%|▌         | 810/14725 [26:01<7:55:05,  2.05s/it]
  6%|▌         | 811/14725 [26:03<7:42:56,  2.00s/it]
  6%|▌         | 812/14725 [26:05<7:32:16,  1.95s/it]
  6%|▌         | 813/14725 [26:07<7:30:04,  1.94s/it]
  6%|▌         | 814/14725 [26:09<7:24:55,  1.92s/it]
  6%|▌         | 815/14725 [26:11<7:26:07,  1.92s/it]
  6%|▌         | 816/14725 [26:13<7:24:08,  1.92s/it]
  6%|▌         | 817/14725 [26:15<7:42:09,  1.99s/it]
+0: {'loss': 0.9321, 'grad_norm': 1.131702581755093, 'learning_rate': 2e-05, 'epoch': 0.06}
+0: {'loss': 0.9364, 'grad_norm': 0.9531283730286756, 'learning_rate': 2e-05, 'epoch': 0.06}
+0:   6%|▌         | 818/14725 [26:17<7:34:56,  1.96s/it]
  6%|▌         | 819/14725 [26:19<7:28:21,  1.93s/it]
  6%|▌         | 820/14725 [26:21<7:26:25,  1.93s/it]
                                                     

  6%|▌         | 820/14725 [26:21<7:26:25,  1.93s/it]
  6%|▌         | 821/14725 [26:22<7:24:35,  1.92s/it]
  6%|▌         | 822/14725 [26:25<7:43:39,  2.00s/it]
  6%|▌         | 823/14725 [26:27<7:33:32,  1.96s/it]
  6%|▌         | 824/14725 [26:29<7:51:08,  2.03s/it]
  6%|▌         | 825/14725 [26:31<7:46:05,  2.01s/it]
  6%|▌         | 826/14725 [26:33<7:35:15,  1.97s/it]
  6%|▌         | 827/14725 [26:34<7:30:24,  1.94s/it]
  6%|▌         | 828/14725 [26:36<7:25:47,  1.92s/it]
  6%|▌         | 829/14725 [26:38<7:26:47,  1.93s/it]
  6%|▌         | 830/14725 [26:40<7:25:16,  1.92s/it]
                                                     

  6%|▌         | 830/14725 [26:40<7:25:16,  1.92s/it]
  6%|▌         | 831/14725 [26:42<7:24:03,  1.92s/it]
  6%|▌         |
+0: {'loss': 0.9239, 'grad_norm': 0.9393252069694837, 'learning_rate': 2e-05, 'epoch': 0.06}
+0:  832/14725 [26:44<7:27:39,  1.93s/it]
  6%|▌         | 833/14725 [26:46<7:21:27,  1.91s/it]
  6%|▌         | 834/14725 [26:48<7:19:01,  1.90s/it]
  6%|▌         | 835/14725 [26:50<7:21:28,  1.91s/it]
  6%|▌         | 836/14725 [26:52<7:28:34,  1.94s/it]
  6%|▌         | 837/14725 [26:54<7:27:07,  1.93s/it]
  6%|▌         | 838/14725 [26:55<7:21:12,  1.91s/it]
  6%|▌         | 839/14725 [26:57<7:24:08,  1.92s/it]
  6%|▌         | 840/14725 [26:59<7:23:33,  1.92s/it]
                                                     

  6%|▌         | 840/14725 [26:59<7:23:33,  1.92s/it]
  6%|▌         | 841/14725 [27:01<7:25:51,  1.93s/it]
  6%|▌         | 842/14725 [27:03<7:22:10,  1.91s/it]
  6%|▌         | 843/14725 [27:05<7:19:44,  1.90s/it]
  6%|▌         | 844/14725 [27:07<7:18:15,  1.89s/it]
  6%|▌         | 845/14725 [27:09<7:13:08,  1.87s/it]
  6%|▌         | 846/14725 [27:11<7:10:38,  1.86s/it]
  6%|▌         | 847/14725 [27:12<7:10:19,  1.86s/it]
  6%|▌         | 848/14725 [27:14
+0: {'loss': 0.929, 'grad_norm': 0.9208762447833763, 'learning_rate': 2e-05, 'epoch': 0.06}
+0: {'loss': 0.93, 'grad_norm': 0.855804311273683, 'learning_rate': 2e-05, 'epoch': 0.06}
+0: <7:13:46,  1.88s/it]
  6%|▌         | 849/14725 [27:16<7:16:31,  1.89s/it]
  6%|▌         | 850/14725 [27:18<7:22:26,  1.91s/it]
                                                     

  6%|▌         | 850/14725 [27:18<7:22:26,  1.91s/it]
  6%|▌         | 851/14725 [27:20<7:27:16,  1.93s/it]
  6%|▌         | 852/14725 [27:22<7:30:45,  1.95s/it]
  6%|▌         | 853/14725 [27:24<7:27:20,  1.93s/it]
  6%|▌         | 854/14725 [27:26<7:25:05,  1.93s/it]
  6%|▌         | 855/14725 [27:28<7:26:47,  1.93s/it]
  6%|▌         | 856/14725 [27:30<7:39:30,  1.99s/it]
  6%|▌         | 857/14725 [27:32<7:36:27,  1.97s/it]
  6%|▌         | 858/14725 [27:34<7:30:52,  1.95s/it]
  6%|▌         | 859/14725 [27:36<7:26:10,  1.93s/it]
  6%|▌         | 860/14725 [27:38<7:48:38,  2.03s/it]
                                                     

  6%|▌         | 860/14725 [27:38<7:48:38,  2.03s/it]
  6%|▌         | 861/14725 [27:40<7:44:11,  2.01s/it]
  6%|▌         | 862/14725 [27:42<7:54:18,  2.05s/i
+0: {'loss': 0.9166, 'grad_norm': 0.8706300411358174, 'learning_rate': 2e-05, 'epoch': 0.06}
+0: t]
  6%|▌         | 863/14725 [27:44<8:10:30,  2.12s/it]
  6%|▌         | 864/14725 [27:47<8:15:51,  2.15s/it]
  6%|▌         | 865/14725 [27:49<8:20:21,  2.17s/it]
  6%|▌         | 866/14725 [27:51<8:24:45,  2.19s/it]
  6%|▌         | 867/14725 [27:53<8:35:02,  2.23s/it]
  6%|▌         | 868/14725 [27:56<8:43:18,  2.27s/it]
  6%|▌         | 869/14725 [27:58<8:22:11,  2.17s/it]
  6%|▌         | 870/14725 [28:00<8:16:09,  2.15s/it]
                                                     

  6%|▌         | 870/14725 [28:00<8:16:09,  2.15s/it]
  6%|▌         | 871/14725 [28:02<8:05:17,  2.10s/it]
  6%|▌         | 872/14725 [28:04<7:52:32,  2.05s/it]
  6%|▌         | 873/14725 [28:06<7:45:48,  2.02s/it]
  6%|▌         | 874/14725 [28:08<8:15:29,  2.15s/it]
  6%|▌         | 875/14725 [28:10<7:59:59,  2.08s/it]
  6%|▌         | 876/14725 [28:12<7:43:31,  2.01s/it]
  6%|▌         | 877/14725 [28:14<7:46:20,  2.02s/it]
  6%|▌         | 878/14725 [28:16<7:35:41,  1.97s/it]
  6%|▌      
+0: {'loss': 0.9254, 'grad_norm': 0.8939325542660427, 'learning_rate': 2e-05, 'epoch': 0.06}
+0: {'loss': 0.9087, 'grad_norm': 0.9101078642215819, 'learning_rate': 2e-05, 'epoch': 0.06}
+0:    | 879/14725 [28:18<7:27:42,  1.94s/it]
  6%|▌         | 880/14725 [28:20<7:32:35,  1.96s/it]
                                                     

  6%|▌         | 880/14725 [28:20<7:32:35,  1.96s/it]
  6%|▌         | 881/14725 [28:22<7:28:16,  1.94s/it]
  6%|▌         | 882/14725 [28:23<7:22:21,  1.92s/it]
  6%|▌         | 883/14725 [28:25<7:24:55,  1.93s/it]
  6%|▌         | 884/14725 [28:27<7:26:25,  1.94s/it]
  6%|▌         | 885/14725 [28:29<7:20:15,  1.91s/it]
  6%|▌         | 886/14725 [28:31<7:46:18,  2.02s/it]
  6%|▌         | 887/14725 [28:34<8:23:38,  2.18s/it]
  6%|▌         | 888/14725 [28:37<9:13:46,  2.40s/it]
  6%|▌         | 889/14725 [28:40<9:58:33,  2.60s/it]
  6%|▌         | 890/14725 [28:42<9:45:36,  2.54s/it]
                                                     

  6%|▌         | 890/14725 [28:42<9:45:36,  2.54s/it]
  6%|▌         | 891/14725 [28:45<9:15:07,  2.41s/it]
  6%|▌         | 892/14725 [28:47<8:49:32,  2.30s/it]
  6%|▌         | 893/14725 [28
+0: {'loss': 0.9213, 'grad_norm': 1.080462267217994, 'learning_rate': 2e-05, 'epoch': 0.06}
+0: :49<9:06:19,  2.37s/it]
  6%|▌         | 894/14725 [28:51<8:57:41,  2.33s/it]
  6%|▌         | 895/14725 [28:53<8:43:14,  2.27s/it]
  6%|▌         | 896/14725 [28:56<8:38:15,  2.25s/it]
  6%|▌         | 897/14725 [28:58<8:13:56,  2.14s/it]
  6%|▌         | 898/14725 [29:00<7:59:30,  2.08s/it]
  6%|▌         | 899/14725 [29:01<7:42:33,  2.01s/it]
  6%|▌         | 900/14725 [29:03<7:31:38,  1.96s/it]
                                                     

  6%|▌         | 900/14725 [29:03<7:31:38,  1.96s/it]
  6%|▌         | 901/14725 [29:05<7:24:40,  1.93s/it]
  6%|▌         | 902/14725 [29:07<7:18:51,  1.90s/it]
  6%|▌         | 903/14725 [29:09<7:12:04,  1.88s/it]
  6%|▌         | 904/14725 [29:10<7:06:17,  1.85s/it]
  6%|▌         | 905/14725 [29:12<7:08:28,  1.86s/it]
  6%|▌         | 906/14725 [29:14<7:04:28,  1.84s/it]
  6%|▌         | 907/14725 [29:17<7:53:03,  2.05s/it]
  6%|▌         | 908/14725 [29:19<8:34:27,  2.23s/it]
  6%|▌         | 909/14725 [29:22<9:07:18,  2.3
+0: {'loss': 0.9281, 'grad_norm': 0.9366175234383184, 'learning_rate': 2e-05, 'epoch': 0.06}
+0: {'loss': 0.9349, 'grad_norm': 0.8872041737066476, 'learning_rate': 2e-05, 'epoch': 0.06}
+0: 8s/it]
  6%|▌         | 910/14725 [29:25<9:27:41,  2.47s/it]
                                                     

  6%|▌         | 910/14725 [29:25<9:27:41,  2.47s/it]
  6%|▌         | 911/14725 [29:27<9:01:55,  2.35s/it]
  6%|▌         | 912/14725 [29:29<8:26:45,  2.20s/it]
  6%|▌         | 913/14725 [29:31<7:59:02,  2.08s/it]
  6%|▌         | 914/14725 [29:32<7:43:39,  2.01s/it]
  6%|▌         | 915/14725 [29:34<7:31:15,  1.96s/it]
  6%|▌         | 916/14725 [29:36<7:24:27,  1.93s/it]
  6%|▌         | 917/14725 [29:38<7:15:45,  1.89s/it]
  6%|▌         | 918/14725 [29:40<7:11:18,  1.87s/it]
  6%|▌         | 919/14725 [29:41<7:04:56,  1.85s/it]
  6%|▌         | 920/14725 [29:43<7:04:51,  1.85s/it]
                                                     

  6%|▌         | 920/14725 [29:43<7:04:51,  1.85s/it]
  6%|▋         | 921/14725 [29:45<7:11:56,  1.88s/it]
  6%|▋         | 922/14725 [29:47<7:08:37,  1.86s/it]
  6%|▋         | 923/14725 [29:49<7:05:06,  1.85s/it]
  6%|▋   
+0: {'loss': 0.9406, 'grad_norm': 0.9332638625758483, 'learning_rate': 2e-05, 'epoch': 0.06}
+0:       | 924/14725 [29:51<7:00:22,  1.83s/it]
  6%|▋         | 925/14725 [29:53<6:59:07,  1.82s/it]
  6%|▋         | 926/14725 [29:54<6:58:45,  1.82s/it]
  6%|▋         | 927/14725 [29:56<6:58:07,  1.82s/it]
  6%|▋         | 928/14725 [29:58<6:56:49,  1.81s/it]
  6%|▋         | 929/14725 [30:00<7:24:34,  1.93s/it]
  6%|▋         | 930/14725 [30:03<8:27:49,  2.21s/it]
                                                     

  6%|▋         | 930/14725 [30:03<8:27:49,  2.21s/it]
  6%|▋         | 931/14725 [30:06<9:31:10,  2.48s/it]
  6%|▋         | 932/14725 [30:09<10:11:28,  2.66s/it]
  6%|▋         | 933/14725 [30:12<10:14:58,  2.68s/it]
  6%|▋         | 934/14725 [30:14<9:15:22,  2.42s/it] 
  6%|▋         | 935/14725 [30:16<8:33:09,  2.23s/it]
  6%|▋         | 936/14725 [30:17<8:03:26,  2.10s/it]
  6%|▋         | 937/14725 [30:19<7:41:29,  2.01s/it]
  6%|▋         | 938/14725 [30:21<7:30:38,  1.96s/it]
  6%|▋         | 939/14725 [30:23<7:18:47,  1.91s/it]
  6%|▋         | 940/14
+0: {'loss': 0.926, 'grad_norm': 0.9419301288837771, 'learning_rate': 2e-05, 'epoch': 0.06}
+0: {'loss': 0.927, 'grad_norm': 0.9246222915589564, 'learning_rate': 2e-05, 'epoch': 0.06}
+0: 725 [30:25<7:09:32,  1.87s/it]
                                                     

  6%|▋         | 940/14725 [30:25<7:09:32,  1.87s/it]
  6%|▋         | 941/14725 [30:26<7:09:16,  1.87s/it]
  6%|▋         | 942/14725 [30:28<7:08:02,  1.86s/it]
  6%|▋         | 943/14725 [30:30<7:04:50,  1.85s/it]
  6%|▋         | 944/14725 [30:32<7:05:18,  1.85s/it]
  6%|▋         | 945/14725 [30:34<7:03:50,  1.85s/it]
  6%|▋         | 946/14725 [30:36<7:00:31,  1.83s/it]
  6%|▋         | 947/14725 [30:37<7:03:10,  1.84s/it]
  6%|▋         | 948/14725 [30:39<6:58:06,  1.82s/it]
  6%|▋         | 949/14725 [30:41<6:58:12,  1.82s/it]
  6%|▋         | 950/14725 [30:43<6:55:18,  1.81s/it]
                                                     

  6%|▋         | 950/14725 [30:43<6:55:18,  1.81s/it]
  6%|▋         | 951/14725 [30:45<7:12:25,  1.88s/it]
  6%|▋         | 952/14725 [30:48<8:11:01,  2.14s/it]
  6%|▋         | 953/14725 [30:50<8:45:00,  2.29s/it]
  6%|▋         | 954/14725 [30:53<9:44:12
+0: {'loss': 0.9197, 'grad_norm': 0.8466817779579806, 'learning_rate': 2e-05, 'epoch': 0.07}
+0: ,  2.55s/it]
  6%|▋         | 955/14725 [30:56<10:10:24,  2.66s/it]
  6%|▋         | 956/14725 [30:58<9:25:39,  2.46s/it] 
  6%|▋         | 957/14725 [31:00<8:38:20,  2.26s/it]
  7%|▋         | 958/14725 [31:02<8:09:58,  2.14s/it]
  7%|▋         | 959/14725 [31:04<7:44:21,  2.02s/it]
  7%|▋         | 960/14725 [31:05<7:26:26,  1.95s/it]
                                                     

  7%|▋         | 960/14725 [31:05<7:26:26,  1.95s/it]
  7%|▋         | 961/14725 [31:07<7:24:07,  1.94s/it]
  7%|▋         | 962/14725 [31:09<7:16:08,  1.90s/it]
  7%|▋         | 963/14725 [31:11<7:08:19,  1.87s/it]
  7%|▋         | 964/14725 [31:13<7:03:44,  1.85s/it]
  7%|▋         | 965/14725 [31:15<6:57:59,  1.82s/it]
  7%|▋         | 966/14725 [31:16<7:00:52,  1.84s/it]
  7%|▋         | 967/14725 [31:18<6:59:45,  1.83s/it]
  7%|▋         | 968/14725 [31:20<7:05:33,  1.86s/it]
  7%|▋         | 969/14725 [31:22<7:00:29,  1.83s/it]
  7%|▋         | 970/14725 [31:24<7:00:56,  1.84s/it]
  
+0: {'loss': 0.9161, 'grad_norm': 0.8459644679474533, 'learning_rate': 2e-05, 'epoch': 0.07}
+0: {'loss': 0.9169, 'grad_norm': 1.0151910116266813, 'learning_rate': 2e-05, 'epoch': 0.07}
+0:                                                    

  7%|▋         | 970/14725 [31:24<7:00:56,  1.84s/it]
  7%|▋         | 971/14725 [31:26<7:02:42,  1.84s/it]
  7%|▋         | 972/14725 [31:27<6:57:05,  1.82s/it]
  7%|▋         | 973/14725 [31:29<7:00:55,  1.84s/it]
  7%|▋         | 974/14725 [31:32<8:06:31,  2.12s/it]
  7%|▋         | 975/14725 [31:35<9:02:20,  2.37s/it]
  7%|▋         | 976/14725 [31:38<9:46:56,  2.56s/it]
  7%|▋         | 977/14725 [31:41<10:08:23,  2.66s/it]
  7%|▋         | 978/14725 [31:43<9:30:04,  2.49s/it] 
  7%|▋         | 979/14725 [31:45<8:46:20,  2.30s/it]
  7%|▋         | 980/14725 [31:47<8:15:02,  2.16s/it]
                                                     

  7%|▋         | 980/14725 [31:47<8:15:02,  2.16s/it]
  7%|▋         | 981/14725 [31:49<7:57:39,  2.09s/it]
  7%|▋         | 982/14725 [31:50<7:37:28,  2.00s/it]
  7%|▋         | 983/14725 [31:52<7:21:52,  1.93s/it]
  7%|▋         | 984/14725 [31:54<7:09:56,  1.88s/it]
  7%|▋         |
+0: {'loss': 0.9287, 'grad_norm': 0.9022180732868995, 'learning_rate': 2e-05, 'epoch': 0.07}
+0:  985/14725 [31:56<7:04:08,  1.85s/it]
  7%|▋         | 986/14725 [31:57<6:59:24,  1.83s/it]
  7%|▋         | 987/14725 [31:59<6:54:47,  1.81s/it]
  7%|▋         | 988/14725 [32:01<6:58:58,  1.83s/it]
  7%|▋         | 989/14725 [32:03<7:03:54,  1.85s/it]
  7%|▋         | 990/14725 [32:05<7:05:04,  1.86s/it]
                                                     

  7%|▋         | 990/14725 [32:05<7:05:04,  1.86s/it]
  7%|▋         | 991/14725 [32:07<7:04:42,  1.86s/it]
  7%|▋         | 992/14725 [32:09<6:59:02,  1.83s/it]
  7%|▋         | 993/14725 [32:10<7:00:07,  1.84s/it]
  7%|▋         | 994/14725 [32:12<6:59:48,  1.83s/it]
  7%|▋         | 995/14725 [32:14<6:56:18,  1.82s/it]
  7%|▋         | 996/14725 [32:16<7:32:00,  1.98s/it]
  7%|▋         | 997/14725 [32:19<8:10:36,  2.14s/it]
  7%|▋         | 998/14725 [32:22<9:11:55,  2.41s/it]
  7%|▋         | 999/14725 [32:25<9:42:43,  2.55s/it]
  7%|▋         | 1000/14725 [32:27<9:20:46,  2.45s/it]
                                  
+0: {'loss': 0.9381, 'grad_norm': 0.8542124372162923, 'learning_rate': 2e-05, 'epoch': 0.07}
+0: {'loss': 0.9115, 'grad_norm': 0.8789449975074507, 'learning_rate': 2e-05, 'epoch': 0.07}
+0:                     

  7%|▋         | 1000/14725 [32:27<9:20:46,  2.45s/it]
  7%|▋         | 1001/14725 [32:29<8:35:23,  2.25s/it]
  7%|▋         | 1002/14725 [32:31<8:03:13,  2.11s/it]
  7%|▋         | 1003/14725 [32:32<7:42:09,  2.02s/it]
  7%|▋         | 1004/14725 [32:34<7:32:06,  1.98s/it]
  7%|▋         | 1005/14725 [32:36<7:23:08,  1.94s/it]
  7%|▋         | 1006/14725 [32:38<7:17:58,  1.92s/it]
  7%|▋         | 1007/14725 [32:40<7:08:18,  1.87s/it]
  7%|▋         | 1008/14725 [32:42<7:04:34,  1.86s/it]
  7%|▋         | 1009/14725 [32:43<7:08:02,  1.87s/it]
  7%|▋         | 1010/14725 [32:45<7:03:24,  1.85s/it]
                                                      

  7%|▋         | 1010/14725 [32:45<7:03:24,  1.85s/it]
  7%|▋         | 1011/14725 [32:47<7:03:00,  1.85s/it]
  7%|▋         | 1012/14725 [32:49<7:04:05,  1.86s/it]
  7%|▋         | 1013/14725 [32:51<6:58:50,  1.83s/it]
  7%|▋         | 1014/14725 [32:53<7:01:46,  1.85s/it]
  7%|▋         | 1015/14725 [32:
+0: {'loss': 0.9226, 'grad_norm': 0.8805726369365436, 'learning_rate': 2e-05, 'epoch': 0.07}
+0: 54<6:57:49,  1.83s/it]
  7%|▋         | 1016/14725 [32:56<7:01:34,  1.85s/it]
  7%|▋         | 1017/14725 [32:58<6:59:25,  1.84s/it]
  7%|▋         | 1018/14725 [33:01<7:40:20,  2.02s/it]
  7%|▋         | 1019/14725 [33:03<8:29:16,  2.23s/it]
  7%|▋         | 1020/14725 [33:06<9:22:17,  2.46s/it]
                                                      

  7%|▋         | 1020/14725 [33:06<9:22:17,  2.46s/it]
  7%|▋         | 1021/14725 [33:09<10:05:31,  2.65s/it]
  7%|▋         | 1022/14725 [33:12<10:08:25,  2.66s/it]
  7%|▋         | 1023/14725 [33:14<9:08:23,  2.40s/it] 
  7%|▋         | 1024/14725 [33:16<8:25:46,  2.21s/it]
  7%|▋         | 1025/14725 [33:17<7:54:47,  2.08s/it]
  7%|▋         | 1026/14725 [33:19<7:41:00,  2.02s/it]
  7%|▋         | 1027/14725 [33:21<7:24:27,  1.95s/it]
  7%|▋         | 1028/14725 [33:23<7:19:22,  1.92s/it]
  7%|▋         | 1029/14725 [33:25<7:11:32,  1.89s/it]
  7%|▋         | 1030/14725 [33:27<7:07:02,  1.87s/it]
                              
+0: {'loss': 0.925, 'grad_norm': 0.9010114397696262, 'learning_rate': 2e-05, 'epoch': 0.07}
+0: {'loss': 0.9324, 'grad_norm': 0.8759380330861883, 'learning_rate': 2e-05, 'epoch': 0.07}
+0:                         

  7%|▋         | 1030/14725 [33:27<7:07:02,  1.87s/it]
  7%|▋         | 1031/14725 [33:28<7:05:42,  1.87s/it]
  7%|▋         | 1032/14725 [33:30<7:03:03,  1.85s/it]
  7%|▋         | 1033/14725 [33:33<8:08:29,  2.14s/it]
  7%|▋         | 1034/14725 [33:35<7:44:18,  2.03s/it]
  7%|▋         | 1035/14725 [33:37<7:34:58,  1.99s/it]
  7%|▋         | 1036/14725 [33:39<7:23:52,  1.95s/it]
  7%|▋         | 1037/14725 [33:40<7:12:16,  1.89s/it]
  7%|▋         | 1038/14725 [33:44<9:08:21,  2.40s/it]
  7%|▋         | 1039/14725 [33:46<9:04:38,  2.39s/it]
  7%|▋         | 1040/14725 [33:49<9:30:47,  2.50s/it]
                                                      

  7%|▋         | 1040/14725 [33:49<9:30:47,  2.50s/it]
  7%|▋         | 1041/14725 [33:52<9:56:15,  2.61s/it]
  7%|▋         | 1042/14725 [33:55<10:40:19,  2.81s/it]
  7%|▋         | 1043/14725 [33:57<9:34:09,  2.52s/it] 
  7%|▋         | 1044/14725 [33:59<8:49:53,  2.32s/it]
  7%|▋         | 1045/1472
+0: {'loss': 0.9175, 'grad_norm': 0.9095463297342189, 'learning_rate': 2e-05, 'epoch': 0.07}
+0: 5 [34:01<8:13:55,  2.17s/it]
  7%|▋         | 1046/14725 [34:03<7:55:41,  2.09s/it]
  7%|▋         | 1047/14725 [34:04<7:38:53,  2.01s/it]
  7%|▋         | 1048/14725 [34:06<7:28:58,  1.97s/it]
  7%|▋         | 1049/14725 [34:08<7:17:34,  1.92s/it]
  7%|▋         | 1050/14725 [34:10<7:23:49,  1.95s/it]
                                                      

  7%|▋         | 1050/14725 [34:10<7:23:49,  1.95s/it]
  7%|▋         | 1051/14725 [34:12<7:19:58,  1.93s/it]
  7%|▋         | 1052/14725 [34:14<7:13:12,  1.90s/it]
  7%|▋         | 1053/14725 [34:16<7:09:16,  1.88s/it]
  7%|▋         | 1054/14725 [34:17<7:02:37,  1.85s/it]
  7%|▋         | 1055/14725 [34:19<7:02:42,  1.86s/it]
  7%|▋         | 1056/14725 [34:21<7:00:41,  1.85s/it]
  7%|▋         | 1057/14725 [34:23<7:04:07,  1.86s/it]
  7%|▋         | 1058/14725 [34:25<7:03:08,  1.86s/it]
  7%|▋         | 1059/14725 [34:27<7:01:29,  1.85s/it]
  7%|▋         | 1060/14725 [34:30<8:09:46,  2.15s/it]
                           
+0: {'loss': 0.9248, 'grad_norm': 0.9323023300894654, 'learning_rate': 2e-05, 'epoch': 0.07}
+0: {'loss': 0.9076, 'grad_norm': 0.8611645943985387, 'learning_rate': 2e-05, 'epoch': 0.07}
+0:                            

  7%|▋         | 1060/14725 [34:30<8:09:46,  2.15s/it]
  7%|▋         | 1061/14725 [34:33<9:04:53,  2.39s/it]
  7%|▋         | 1062/14725 [34:35<9:18:02,  2.45s/it]
  7%|▋         | 1063/14725 [34:37<8:33:55,  2.26s/it]
  7%|▋         | 1064/14725 [34:39<8:05:38,  2.13s/it]
  7%|▋         | 1065/14725 [34:41<7:45:51,  2.05s/it]
  7%|▋         | 1066/14725 [34:42<7:31:47,  1.98s/it]
  7%|▋         | 1067/14725 [34:44<7:22:37,  1.94s/it]
  7%|▋         | 1068/14725 [34:46<7:14:31,  1.91s/it]
  7%|▋         | 1069/14725 [34:48<7:09:26,  1.89s/it]
  7%|▋         | 1070/14725 [34:50<7:07:15,  1.88s/it]
                                                      

  7%|▋         | 1070/14725 [34:50<7:07:15,  1.88s/it]
  7%|▋         | 1071/14725 [34:52<7:00:48,  1.85s/it]
  7%|▋         | 1072/14725 [34:53<6:55:31,  1.83s/it]
  7%|▋         | 1073/14725 [34:55<6:51:42,  1.81s/it]
  7%|▋         | 1074/14725 [34:57<6:49:57,  1.80s/it]
  7%|▋         | 1075/147
+0: {'loss': 0.9157, 'grad_norm': 0.8239568345748762, 'learning_rate': 2e-05, 'epoch': 0.07}
+0: 25 [34:59<6:52:16,  1.81s/it]
  7%|▋         | 1076/14725 [35:01<7:05:52,  1.87s/it]
  7%|▋         | 1077/14725 [35:03<6:59:43,  1.85s/it]
  7%|▋         | 1078/14725 [35:04<6:53:34,  1.82s/it]
  7%|▋         | 1079/14725 [35:06<6:56:05,  1.83s/it]
  7%|▋         | 1080/14725 [35:09<7:42:03,  2.03s/it]
                                                      

  7%|▋         | 1080/14725 [35:09<7:42:03,  2.03s/it]
  7%|▋         | 1081/14725 [35:12<8:43:28,  2.30s/it]
  7%|▋         | 1082/14725 [35:15<9:24:05,  2.48s/it]
  7%|▋         | 1083/14725 [35:16<8:38:11,  2.28s/it]
  7%|▋         | 1084/14725 [35:18<8:14:48,  2.18s/it]
  7%|▋         | 1085/14725 [35:20<7:50:23,  2.07s/it]
  7%|▋         | 1086/14725 [35:22<7:32:29,  1.99s/it]
  7%|▋         | 1087/14725 [35:24<7:27:07,  1.97s/it]
  7%|▋         | 1088/14725 [35:26<7:19:10,  1.93s/it]
  7%|▋         | 1089/14725 [35:28<7:13:28,  1.91s/it]
  7%|▋         | 1090/14725 [35:29<7:09:28,  1.89s/it]
                          
+0: {'loss': 0.9133, 'grad_norm': 0.866106096396927, 'learning_rate': 2e-05, 'epoch': 0.07}
+0: {'loss': 0.9211, 'grad_norm': 0.852100038179294, 'learning_rate': 2e-05, 'epoch': 0.07}
+0:                             

  7%|▋         | 1090/14725 [35:29<7:09:28,  1.89s/it]
  7%|▋         | 1091/14725 [35:31<7:04:25,  1.87s/it]
  7%|▋         | 1092/14725 [35:33<6:58:21,  1.84s/it]
  7%|▋         | 1093/14725 [35:35<6:53:51,  1.82s/it]
  7%|▋         | 1094/14725 [35:37<6:57:13,  1.84s/it]
  7%|▋         | 1095/14725 [35:38<6:57:02,  1.84s/it]
  7%|▋         | 1096/14725 [35:40<6:57:38,  1.84s/it]
  7%|▋         | 1097/14725 [35:42<6:57:48,  1.84s/it]
  7%|▋         | 1098/14725 [35:44<7:01:09,  1.85s/it]
  7%|▋         | 1099/14725 [35:46<6:56:31,  1.83s/it]
  7%|▋         | 1100/14725 [35:48<7:33:52,  2.00s/it]
                                                      

  7%|▋         | 1100/14725 [35:48<7:33:52,  2.00s/it]
  7%|▋         | 1101/14725 [35:51<8:44:01,  2.31s/it]
  7%|▋         | 1102/14725 [35:54<9:26:32,  2.50s/it]
  7%|▋         | 1103/14725 [35:56<8:50:10,  2.34s/it]
  7%|▋         | 1104/14725 [35:58<8:14:41,  2.18s/it]
  8%|▊         | 1105/14
+0: {'loss': 0.9261, 'grad_norm': 1.0014343608776637, 'learning_rate': 2e-05, 'epoch': 0.08}
+0: 725 [36:00<7:53:37,  2.09s/it]
  8%|▊         | 1106/14725 [36:02<7:32:10,  1.99s/it]
  8%|▊         | 1107/14725 [36:03<7:22:03,  1.95s/it]
  8%|▊         | 1108/14725 [36:05<7:11:58,  1.90s/it]
  8%|▊         | 1109/14725 [36:07<7:06:45,  1.88s/it]
  8%|▊         | 1110/14725 [36:09<7:01:53,  1.86s/it]
                                                      

  8%|▊         | 1110/14725 [36:09<7:01:53,  1.86s/it]
  8%|▊         | 1111/14725 [36:11<7:00:21,  1.85s/it]
  8%|▊         | 1112/14725 [36:12<6:56:32,  1.84s/it]
  8%|▊         | 1113/14725 [36:14<6:56:40,  1.84s/it]
  8%|▊         | 1114/14725 [36:16<6:58:40,  1.85s/it]
  8%|▊         | 1115/14725 [36:18<6:56:29,  1.84s/it]
  8%|▊         | 1116/14725 [36:20<6:54:38,  1.83s/it]
  8%|▊         | 1117/14725 [36:22<6:57:23,  1.84s/it]
  8%|▊         | 1118/14725 [36:24<6:56:55,  1.84s/it]
  8%|▊         | 1119/14725 [36:25<6:58:52,  1.85s/it]
  8%|▊         | 1120/14725 [36:27<6:59:16,  1.85s/it]
                         
+0: {'loss': 0.9273, 'grad_norm': 0.9639010464512076, 'learning_rate': 2e-05, 'epoch': 0.08}
+0: {'loss': 0.9107, 'grad_norm': 0.9035199098411604, 'learning_rate': 2e-05, 'epoch': 0.08}
+0:                              

  8%|▊         | 1120/14725 [36:27<6:59:16,  1.85s/it]
  8%|▊         | 1121/14725 [36:30<8:06:18,  2.14s/it]
  8%|▊         | 1122/14725 [36:33<9:02:53,  2.39s/it]
  8%|▊         | 1123/14725 [36:35<8:59:27,  2.38s/it]
  8%|▊         | 1124/14725 [36:37<8:21:25,  2.21s/it]
  8%|▊         | 1125/14725 [36:39<7:52:36,  2.09s/it]
  8%|▊         | 1126/14725 [36:41<7:35:51,  2.01s/it]
  8%|▊         | 1127/14725 [36:43<7:25:08,  1.96s/it]
  8%|▊         | 1128/14725 [36:45<7:14:29,  1.92s/it]
  8%|▊         | 1129/14725 [36:46<7:07:43,  1.89s/it]
  8%|▊         | 1130/14725 [36:48<7:05:58,  1.88s/it]
                                                      

  8%|▊         | 1130/14725 [36:48<7:05:58,  1.88s/it]
  8%|▊         | 1131/14725 [36:50<7:04:38,  1.87s/it]
  8%|▊         | 1132/14725 [36:52<7:01:17,  1.86s/it]
  8%|▊         | 1133/14725 [36:54<6:58:48,  1.85s/it]
  8%|▊         | 1134/14725 [36:55<6:55:03,  1.83s/it]
  8%|▊         | 1135/1
+0: {'loss': 0.9103, 'grad_norm': 1.0157013182266004, 'learning_rate': 2e-05, 'epoch': 0.08}
+0: 4725 [36:57<7:01:58,  1.86s/it]
  8%|▊         | 1136/14725 [36:59<6:55:26,  1.83s/it]
  8%|▊         | 1137/14725 [37:01<7:01:15,  1.86s/it]
  8%|▊         | 1138/14725 [37:03<7:01:58,  1.86s/it]
  8%|▊         | 1139/14725 [37:05<6:57:14,  1.84s/it]
  8%|▊         | 1140/14725 [37:07<6:53:08,  1.82s/it]
                                                      

  8%|▊         | 1140/14725 [37:07<6:53:08,  1.82s/it]
  8%|▊         | 1141/14725 [37:09<7:40:37,  2.03s/it]
  8%|▊         | 1142/14725 [37:12<8:52:08,  2.35s/it]
  8%|▊         | 1143/14725 [37:15<9:24:55,  2.50s/it]
  8%|▊         | 1144/14725 [37:17<8:35:37,  2.28s/it]
  8%|▊         | 1145/14725 [37:19<8:00:49,  2.12s/it]
  8%|▊         | 1146/14725 [37:20<7:37:19,  2.02s/it]
  8%|▊         | 1147/14725 [37:22<7:20:34,  1.95s/it]
  8%|▊         | 1148/14725 [37:24<7:10:49,  1.90s/it]
  8%|▊         | 1149/14725 [37:26<7:03:12,  1.87s/it]
  8%|▊         | 1150/14725 [37:28<7:00:19,  1.86s/it]
                        
+0: {'loss': 0.9145, 'grad_norm': 0.8475864069132946, 'learning_rate': 2e-05, 'epoch': 0.08}
+0: {'loss': 0.9238, 'grad_norm': 0.9101081831517239, 'learning_rate': 2e-05, 'epoch': 0.08}
+0:                               

  8%|▊         | 1150/14725 [37:28<7:00:19,  1.86s/it]
  8%|▊         | 1151/14725 [37:29<6:58:04,  1.85s/it]
  8%|▊         | 1152/14725 [37:31<6:52:51,  1.83s/it]
  8%|▊         | 1153/14725 [37:33<6:58:13,  1.85s/it]
  8%|▊         | 1154/14725 [37:35<6:56:26,  1.84s/it]
  8%|▊         | 1155/14725 [37:37<6:56:18,  1.84s/it]
  8%|▊         | 1156/14725 [37:38<6:52:47,  1.83s/it]
  8%|▊         | 1157/14725 [37:40<6:56:31,  1.84s/it]
  8%|▊         | 1158/14725 [37:42<6:54:52,  1.83s/it]
  8%|▊         | 1159/14725 [37:44<7:01:43,  1.87s/it]
  8%|▊         | 1160/14725 [37:46<7:00:47,  1.86s/it]
                                                      

  8%|▊         | 1160/14725 [37:46<7:00:47,  1.86s/it]
  8%|▊         | 1161/14725 [37:48<7:16:23,  1.93s/it]
  8%|▊         | 1162/14725 [37:51<8:25:44,  2.24s/it]
  8%|▊         | 1163/14725 [37:54<8:53:53,  2.36s/it]
  8%|▊         | 1164/14725 [37:56<8:26:52,  2.24s/it]
  8%|▊         | 1165/
+0: {'loss': 0.9115, 'grad_norm': 0.8882528951663808, 'learning_rate': 2e-05, 'epoch': 0.08}
+0: 14725 [37:57<7:59:36,  2.12s/it]
  8%|▊         | 1166/14725 [37:59<7:38:20,  2.03s/it]
  8%|▊         | 1167/14725 [38:01<7:28:40,  1.99s/it]
  8%|▊         | 1168/14725 [38:03<7:20:42,  1.95s/it]
  8%|▊         | 1169/14725 [38:05<7:08:39,  1.90s/it]
  8%|▊         | 1170/14725 [38:07<7:00:52,  1.86s/it]
                                                      

  8%|▊         | 1170/14725 [38:07<7:00:52,  1.86s/it]
  8%|▊         | 1171/14725 [38:08<6:56:51,  1.85s/it]
  8%|▊         | 1172/14725 [38:10<6:57:13,  1.85s/it]
  8%|▊         | 1173/14725 [38:12<6:50:47,  1.82s/it]
  8%|▊         | 1174/14725 [38:14<6:51:28,  1.82s/it]
  8%|▊         | 1175/14725 [38:16<6:48:46,  1.81s/it]
  8%|▊         | 1176/14725 [38:17<6:49:48,  1.81s/it]
  8%|▊         | 1177/14725 [38:19<6:46:15,  1.80s/it]
  8%|▊         | 1178/14725 [38:21<6:53:08,  1.83s/it]
  8%|▊         | 1179/14725 [38:23<7:02:11,  1.87s/it]
  8%|▊         | 1180/14725 [38:25<6:57:18,  1.85s/it]
                       
+0: {'loss': 0.929, 'grad_norm': 0.9531580679656779, 'learning_rate': 2e-05, 'epoch': 0.08}
+0: {'loss': 0.9351, 'grad_norm': 0.7899421394616277, 'learning_rate': 2e-05, 'epoch': 0.08}
+0:                                

  8%|▊         | 1180/14725 [38:25<6:57:18,  1.85s/it]
  8%|▊         | 1181/14725 [38:27<6:57:02,  1.85s/it]
  8%|▊         | 1182/14725 [38:29<7:39:44,  2.04s/it]
  8%|▊         | 1183/14725 [38:32<8:40:35,  2.31s/it]
  8%|▊         | 1184/14725 [38:35<8:55:04,  2.37s/it]
  8%|▊         | 1185/14725 [38:36<8:15:05,  2.19s/it]
  8%|▊         | 1186/14725 [38:38<7:47:47,  2.07s/it]
  8%|▊         | 1187/14725 [38:40<7:27:17,  1.98s/it]
  8%|▊         | 1188/14725 [38:43<8:39:38,  2.30s/it]
  8%|▊         | 1189/14725 [38:47<10:16:46,  2.73s/it]
  8%|▊         | 1190/14725 [38:49<9:57:32,  2.65s/it] 
                                                      

  8%|▊         | 1190/14725 [38:49<9:57:32,  2.65s/it]
  8%|▊         | 1191/14725 [38:51<9:32:26,  2.54s/it]
  8%|▊         | 1192/14725 [38:53<8:50:53,  2.35s/it]
  8%|▊         | 1193/14725 [38:55<8:13:31,  2.19s/it]
  8%|▊         | 1194/14725 [38:57<7:49:55,  2.08s/it]
  8%|▊         | 11
+0: {'loss': 0.8908, 'grad_norm': 0.8366564182466905, 'learning_rate': 2e-05, 'epoch': 0.08}
+0: 95/14725 [38:59<7:34:15,  2.01s/it]
  8%|▊         | 1196/14725 [39:01<7:17:33,  1.94s/it]
  8%|▊         | 1197/14725 [39:03<7:19:47,  1.95s/it]
  8%|▊         | 1198/14725 [39:05<7:15:57,  1.93s/it]
  8%|▊         | 1199/14725 [39:07<7:19:56,  1.95s/it]
  8%|▊         | 1200/14725 [39:09<7:39:46,  2.04s/it]
                                                      

  8%|▊         | 1200/14725 [39:09<7:39:46,  2.04s/it]
  8%|▊         | 1201/14725 [39:11<7:26:41,  1.98s/it]
  8%|▊         | 1202/14725 [39:13<7:20:12,  1.95s/it]
  8%|▊         | 1203/14725 [39:14<7:15:58,  1.93s/it]
  8%|▊         | 1204/14725 [39:16<7:06:57,  1.89s/it]
  8%|▊         | 1205/14725 [39:18<7:01:06,  1.87s/it]
  8%|▊         | 1206/14725 [39:20<6:58:10,  1.86s/it]
  8%|▊         | 1207/14725 [39:22<6:56:50,  1.85s/it]
  8%|▊         | 1208/14725 [39:24<6:55:40,  1.85s/it]
  8%|▊         | 1209/14725 [39:25<7:00:28,  1.87s/it]
  8%|▊         | 1210/14725 [39:27<6:55:16,  1.84s/it]
                    
+0: {'loss': 0.8963, 'grad_norm': 0.8384489031158491, 'learning_rate': 2e-05, 'epoch': 0.08}
+0: {'loss': 0.9235, 'grad_norm': 0.9070688791504693, 'learning_rate': 2e-05, 'epoch': 0.08}
+0:                                   

  8%|▊         | 1210/14725 [39:27<6:55:16,  1.84s/it]
  8%|▊         | 1211/14725 [39:29<6:52:46,  1.83s/it]
  8%|▊         | 1212/14725 [39:31<6:53:38,  1.84s/it]
  8%|▊         | 1213/14725 [39:33<6:53:31,  1.84s/it]
  8%|▊         | 1214/14725 [39:35<7:01:24,  1.87s/it]
  8%|▊         | 1215/14725 [39:37<7:09:39,  1.91s/it]
  8%|▊         | 1216/14725 [39:39<7:24:21,  1.97s/it]
  8%|▊         | 1217/14725 [39:41<7:21:42,  1.96s/it]
  8%|▊         | 1218/14725 [39:43<7:13:28,  1.93s/it]
  8%|▊         | 1219/14725 [39:44<7:04:39,  1.89s/it]
  8%|▊         | 1220/14725 [39:46<7:00:18,  1.87s/it]
                                                      

  8%|▊         | 1220/14725 [39:46<7:00:18,  1.87s/it]
  8%|▊         | 1221/14725 [39:48<6:56:18,  1.85s/it]
  8%|▊         | 1222/14725 [39:50<6:55:32,  1.85s/it]
  8%|▊         | 1223/14725 [39:53<7:56:43,  2.12s/it]
  8%|▊         | 1224/14725 [39:54<7:35:13,  2.02s/it]
  8%|▊         | 1
+0: {'loss': 0.9024, 'grad_norm': 0.9520334832022253, 'learning_rate': 2e-05, 'epoch': 0.08}
+0: 225/14725 [39:56<7:22:29,  1.97s/it]
  8%|▊         | 1226/14725 [39:58<7:09:21,  1.91s/it]
  8%|▊         | 1227/14725 [40:00<7:04:40,  1.89s/it]
  8%|▊         | 1228/14725 [40:02<7:05:02,  1.89s/it]
  8%|▊         | 1229/14725 [40:04<6:58:09,  1.86s/it]
  8%|▊         | 1230/14725 [40:05<6:53:22,  1.84s/it]
                                                      

  8%|▊         | 1230/14725 [40:05<6:53:22,  1.84s/it]
  8%|▊         | 1231/14725 [40:07<6:49:52,  1.82s/it]
  8%|▊         | 1232/14725 [40:09<6:48:21,  1.82s/it]
  8%|▊         | 1233/14725 [40:11<6:46:21,  1.81s/it]
  8%|▊         | 1234/14725 [40:13<6:47:38,  1.81s/it]
  8%|▊         | 1235/14725 [40:14<6:50:01,  1.82s/it]
  8%|▊         | 1236/14725 [40:16<6:47:27,  1.81s/it]
  8%|▊         | 1237/14725 [40:18<6:46:20,  1.81s/it]
  8%|▊         | 1238/14725 [40:20<6:49:32,  1.82s/it]
  8%|▊         | 1239/14725 [40:22<6:50:32,  1.83s/it]
  8%|▊         | 1240/14725 [40:23<6:47:36,  1.81s/it]
                   
+0: {'loss': 0.8832, 'grad_norm': 0.9350088240550058, 'learning_rate': 2e-05, 'epoch': 0.08}
+0: {'loss': 0.8982, 'grad_norm': 0.9420011744973924, 'learning_rate': 2e-05, 'epoch': 0.08}
+0:                                    

  8%|▊         | 1240/14725 [40:23<6:47:36,  1.81s/it]
  8%|▊         | 1241/14725 [40:25<6:54:49,  1.85s/it]
  8%|▊         | 1242/14725 [40:27<6:53:00,  1.84s/it]
  8%|▊         | 1243/14725 [40:29<6:53:42,  1.84s/it]
  8%|▊         | 1244/14725 [40:31<6:50:23,  1.83s/it]
  8%|▊         | 1245/14725 [40:33<6:46:53,  1.81s/it]
  8%|▊         | 1246/14725 [40:34<6:47:56,  1.82s/it]
  8%|▊         | 1247/14725 [40:36<6:45:59,  1.81s/it]
  8%|▊         | 1248/14725 [40:38<6:46:23,  1.81s/it]
  8%|▊         | 1249/14725 [40:40<6:47:33,  1.81s/it]
  8%|▊         | 1250/14725 [40:42<6:43:19,  1.80s/it]
                                                      

  8%|▊         | 1250/14725 [40:42<6:43:19,  1.80s/it]
  8%|▊         | 1251/14725 [40:43<6:44:01,  1.80s/it]
  9%|▊         | 1252/14725 [40:45<6:45:52,  1.81s/it]
  9%|▊         | 1253/14725 [40:47<6:49:18,  1.82s/it]
  9%|▊         | 1254/14725 [40:49<6:47:38,  1.82s/it]
  9%|▊         | 
+0: {'loss': 0.9162, 'grad_norm': 0.9500709320854123, 'learning_rate': 2e-05, 'epoch': 0.09}
+0: 1255/14725 [40:51<6:45:44,  1.81s/it]
  9%|▊         | 1256/14725 [40:52<6:48:08,  1.82s/it]
  9%|▊         | 1257/14725 [40:54<6:54:06,  1.84s/it]
  9%|▊         | 1258/14725 [40:56<6:49:25,  1.82s/it]
  9%|▊         | 1259/14725 [40:58<6:48:22,  1.82s/it]
  9%|▊         | 1260/14725 [41:00<6:50:38,  1.83s/it]
                                                      

  9%|▊         | 1260/14725 [41:00<6:50:38,  1.83s/it]
  9%|▊         | 1261/14725 [41:02<6:47:34,  1.82s/it]
  9%|▊         | 1262/14725 [41:03<6:45:55,  1.81s/it]
  9%|▊         | 1263/14725 [41:05<6:43:02,  1.80s/it]
  9%|▊         | 1264/14725 [41:07<6:42:21,  1.79s/it]
  9%|▊         | 1265/14725 [41:09<6:43:38,  1.80s/it]
  9%|▊         | 1266/14725 [41:11<6:43:40,  1.80s/it]
  9%|▊         | 1267/14725 [41:12<6:46:52,  1.81s/it]
  9%|▊         | 1268/14725 [41:14<6:46:58,  1.81s/it]
  9%|▊         | 1269/14725 [41:16<6:43:30,  1.80s/it]
  9%|▊         | 1270/14725 [41:18<6:51:12,  1.83s/it]
                  
+0: {'loss': 0.8942, 'grad_norm': 0.8714804339868751, 'learning_rate': 2e-05, 'epoch': 0.09}
+0: {'loss': 0.8848, 'grad_norm': 0.8586487041037866, 'learning_rate': 2e-05, 'epoch': 0.09}
+0:                                     

  9%|▊         | 1270/14725 [41:18<6:51:12,  1.83s/it]
  9%|▊         | 1271/14725 [41:20<6:51:14,  1.83s/it]
  9%|▊         | 1272/14725 [41:22<6:52:29,  1.84s/it]
  9%|▊         | 1273/14725 [41:23<6:51:31,  1.84s/it]
  9%|▊         | 1274/14725 [41:25<6:46:42,  1.81s/it]
  9%|▊         | 1275/14725 [41:27<6:45:24,  1.81s/it]
  9%|▊         | 1276/14725 [41:29<6:42:32,  1.80s/it]
  9%|▊         | 1277/14725 [41:31<6:45:03,  1.81s/it]
  9%|▊         | 1278/14725 [41:32<6:45:55,  1.81s/it]
  9%|▊         | 1279/14725 [41:34<6:46:14,  1.81s/it]
  9%|▊         | 1280/14725 [41:36<6:46:08,  1.81s/it]
                                                      

  9%|▊         | 1280/14725 [41:36<6:46:08,  1.81s/it]
  9%|▊         | 1281/14725 [41:38<6:50:41,  1.83s/it]
  9%|▊         | 1282/14725 [41:40<6:51:02,  1.83s/it]
  9%|▊         | 1283/14725 [41:42<6:53:46,  1.85s/it]
  9%|▊         | 1284/14725 [41:43<6:53:00,  1.84s/it]
  9%|▊         |
+0: {'loss': 0.9159, 'grad_norm': 0.8951412919235223, 'learning_rate': 2e-05, 'epoch': 0.09}
+0:  1285/14725 [41:45<6:53:10,  1.84s/it]
  9%|▊         | 1286/14725 [41:47<6:52:34,  1.84s/it]
  9%|▊         | 1287/14725 [41:49<6:50:49,  1.83s/it]
  9%|▊         | 1288/14725 [41:51<6:48:23,  1.82s/it]
  9%|▉         | 1289/14725 [41:53<6:52:37,  1.84s/it]
  9%|▉         | 1290/14725 [41:54<6:49:19,  1.83s/it]
                                                      

  9%|▉         | 1290/14725 [41:54<6:49:19,  1.83s/it]
  9%|▉         | 1291/14725 [41:56<6:48:24,  1.82s/it]
  9%|▉         | 1292/14725 [41:58<6:47:12,  1.82s/it]
  9%|▉         | 1293/14725 [42:00<6:45:04,  1.81s/it]
  9%|▉         | 1294/14725 [42:02<6:45:41,  1.81s/it]
  9%|▉         | 1295/14725 [42:04<6:47:33,  1.82s/it]
  9%|▉         | 1296/14725 [42:05<6:43:06,  1.80s/it]
  9%|▉         | 1297/14725 [42:07<6:41:50,  1.80s/it]
  9%|▉         | 1298/14725 [42:09<6:44:39,  1.81s/it]
  9%|▉         | 1299/14725 [42:11<6:42:55,  1.80s/it]
  9%|▉         | 1300/14725 [42:12<6:43:10,  1.80s/it]
                 
+0: {'loss': 0.8994, 'grad_norm': 0.8742586600522119, 'learning_rate': 2e-05, 'epoch': 0.09}
+0: {'loss': 0.9157, 'grad_norm': 0.9216501753557917, 'learning_rate': 2e-05, 'epoch': 0.09}
+0:                                      

  9%|▉         | 1300/14725 [42:12<6:43:10,  1.80s/it]
  9%|▉         | 1301/14725 [42:14<6:43:42,  1.80s/it]
  9%|▉         | 1302/14725 [42:16<6:44:34,  1.81s/it]
  9%|▉         | 1303/14725 [42:18<6:45:02,  1.81s/it]
  9%|▉         | 1304/14725 [42:20<6:50:41,  1.84s/it]
  9%|▉         | 1305/14725 [42:22<6:50:37,  1.84s/it]
  9%|▉         | 1306/14725 [42:23<6:46:36,  1.82s/it]
  9%|▉         | 1307/14725 [42:25<6:48:14,  1.83s/it]
  9%|▉         | 1308/14725 [42:27<6:48:06,  1.83s/it]
  9%|▉         | 1309/14725 [42:29<6:47:24,  1.82s/it]
  9%|▉         | 1310/14725 [42:31<6:45:32,  1.81s/it]
                                                      

  9%|▉         | 1310/14725 [42:31<6:45:32,  1.81s/it]
  9%|▉         | 1311/14725 [42:33<6:43:56,  1.81s/it]
  9%|▉         | 1312/14725 [42:34<6:47:39,  1.82s/it]
  9%|▉         | 1313/14725 [42:36<6:43:09,  1.80s/it]
  9%|▉         | 1314/14725 [42:38<6:44:09,  1.81s/it]
  9%|▉         
+0: {'loss': 0.9049, 'grad_norm': 0.8268809896376836, 'learning_rate': 2e-05, 'epoch': 0.09}
+0: | 1315/14725 [42:40<6:45:07,  1.81s/it]
  9%|▉         | 1316/14725 [42:42<6:44:01,  1.81s/it]
  9%|▉         | 1317/14725 [42:43<6:44:20,  1.81s/it]
  9%|▉         | 1318/14725 [42:45<6:43:59,  1.81s/it]
  9%|▉         | 1319/14725 [42:47<6:44:12,  1.81s/it]
  9%|▉         | 1320/14725 [42:49<6:43:35,  1.81s/it]
                                                      

  9%|▉         | 1320/14725 [42:49<6:43:35,  1.81s/it]
  9%|▉         | 1321/14725 [42:51<6:50:47,  1.84s/it]
  9%|▉         | 1322/14725 [42:53<6:52:06,  1.84s/it]
  9%|▉         | 1323/14725 [42:54<6:48:04,  1.83s/it]
  9%|▉         | 1324/14725 [42:56<6:44:42,  1.81s/it]
  9%|▉         | 1325/14725 [42:58<6:48:14,  1.83s/it]
  9%|▉         | 1326/14725 [43:00<6:48:28,  1.83s/it]
  9%|▉         | 1327/14725 [43:02<7:02:52,  1.89s/it]
  9%|▉         | 1328/14725 [43:04<7:02:41,  1.89s/it]
  9%|▉         | 1329/14725 [43:06<6:58:17,  1.87s/it]
  9%|▉         | 1330/14725 [43:07<6:55:54,  1.86s/it]
                
+0: {'loss': 0.9047, 'grad_norm': 0.9317500261634566, 'learning_rate': 2e-05, 'epoch': 0.09}
+0: {'loss': 0.9059, 'grad_norm': 0.9244855455253193, 'learning_rate': 2e-05, 'epoch': 0.09}
+0:                                       

  9%|▉         | 1330/14725 [43:07<6:55:54,  1.86s/it]
  9%|▉         | 1331/14725 [43:09<6:51:56,  1.85s/it]
  9%|▉         | 1332/14725 [43:11<6:48:00,  1.83s/it]
  9%|▉         | 1333/14725 [43:13<6:46:58,  1.82s/it]
  9%|▉         | 1334/14725 [43:15<6:46:51,  1.82s/it]
  9%|▉         | 1335/14725 [43:17<6:48:27,  1.83s/it]
  9%|▉         | 1336/14725 [43:18<6:44:34,  1.81s/it]
  9%|▉         | 1337/14725 [43:20<6:48:18,  1.83s/it]
  9%|▉         | 1338/14725 [43:22<6:46:55,  1.82s/it]
  9%|▉         | 1339/14725 [43:24<6:43:28,  1.81s/it]
  9%|▉         | 1340/14725 [43:26<6:42:01,  1.80s/it]
                                                      

  9%|▉         | 1340/14725 [43:26<6:42:01,  1.80s/it]
  9%|▉         | 1341/14725 [43:27<6:42:43,  1.81s/it]
  9%|▉         | 1342/14725 [43:29<6:42:29,  1.80s/it]
  9%|▉         | 1343/14725 [43:31<6:39:45,  1.79s/it]
  9%|▉         | 1344/14725 [43:33<6:44:29,  1.81s/it]
  9%|▉        
+0: {'loss': 0.9059, 'grad_norm': 0.8191216843468432, 'learning_rate': 2e-05, 'epoch': 0.09}
+0:  | 1345/14725 [43:35<6:43:48,  1.81s/it]
  9%|▉         | 1346/14725 [43:36<6:42:46,  1.81s/it]
  9%|▉         | 1347/14725 [43:38<6:45:50,  1.82s/it]
  9%|▉         | 1348/14725 [43:40<6:42:35,  1.81s/it]
  9%|▉         | 1349/14725 [43:42<6:44:24,  1.81s/it]
  9%|▉         | 1350/14725 [43:44<6:46:52,  1.83s/it]
                                                      

  9%|▉         | 1350/14725 [43:44<6:46:52,  1.83s/it]
  9%|▉         | 1351/14725 [43:45<6:45:06,  1.82s/it]
  9%|▉         | 1352/14725 [43:47<6:48:31,  1.83s/it]
  9%|▉         | 1353/14725 [43:49<6:49:14,  1.84s/it]
  9%|▉         | 1354/14725 [43:51<6:50:28,  1.84s/it]
  9%|▉         | 1355/14725 [43:53<6:50:13,  1.84s/it]
  9%|▉         | 1356/14725 [43:55<6:50:46,  1.84s/it]
  9%|▉         | 1357/14725 [43:57<6:48:52,  1.84s/it]
  9%|▉         | 1358/14725 [43:58<6:51:12,  1.85s/it]
  9%|▉         | 1359/14725 [44:00<6:55:31,  1.87s/it]
  9%|▉         | 1360/14725 [44:02<6:52:58,  1.85s/it]
               
+0: {'loss': 0.9168, 'grad_norm': 0.8223899729954157, 'learning_rate': 2e-05, 'epoch': 0.09}
+0: {'loss': 0.9246, 'grad_norm': 0.8128996439710463, 'learning_rate': 2e-05, 'epoch': 0.09}
+0:                                        

  9%|▉         | 1360/14725 [44:02<6:52:58,  1.85s/it]
  9%|▉         | 1361/14725 [44:04<6:50:38,  1.84s/it]
  9%|▉         | 1362/14725 [44:06<6:48:23,  1.83s/it]
  9%|▉         | 1363/14725 [44:08<6:47:58,  1.83s/it]
  9%|▉         | 1364/14725 [44:09<6:45:26,  1.82s/it]
  9%|▉         | 1365/14725 [44:11<6:43:49,  1.81s/it]
  9%|▉         | 1366/14725 [44:13<6:55:06,  1.86s/it]
  9%|▉         | 1367/14725 [44:15<6:50:44,  1.84s/it]
  9%|▉         | 1368/14725 [44:17<6:49:42,  1.84s/it]
  9%|▉         | 1369/14725 [44:19<6:53:21,  1.86s/it]
  9%|▉         | 1370/14725 [44:21<6:51:08,  1.85s/it]
                                                      

  9%|▉         | 1370/14725 [44:21<6:51:08,  1.85s/it]
  9%|▉         | 1371/14725 [44:22<6:47:38,  1.83s/it]
  9%|▉         | 1372/14725 [44:24<6:44:22,  1.82s/it]
  9%|▉         | 1373/14725 [44:26<6:41:54,  1.81s/it]
  9%|▉         | 1374/14725 [44:28<6:44:24,  1.82s/it]
  9%|▉       
+0: {'loss': 0.8686, 'grad_norm': 0.9191740643219131, 'learning_rate': 2e-05, 'epoch': 0.09}
+0:   | 1375/14725 [44:30<6:46:34,  1.83s/it]
  9%|▉         | 1376/14725 [44:31<6:43:12,  1.81s/it]
  9%|▉         | 1377/14725 [44:33<6:46:45,  1.83s/it]
  9%|▉         | 1378/14725 [44:35<6:43:51,  1.82s/it]
  9%|▉         | 1379/14725 [44:37<6:40:07,  1.80s/it]
  9%|▉         | 1380/14725 [44:39<6:43:06,  1.81s/it]
                                                      

  9%|▉         | 1380/14725 [44:39<6:43:06,  1.81s/it]
  9%|▉         | 1381/14725 [44:40<6:45:54,  1.83s/it]
  9%|▉         | 1382/14725 [44:42<6:44:51,  1.82s/it]
  9%|▉         | 1383/14725 [44:44<6:43:37,  1.82s/it]
  9%|▉         | 1384/14725 [44:46<6:48:45,  1.84s/it]
  9%|▉         | 1385/14725 [44:48<6:50:03,  1.84s/it]
  9%|▉         | 1386/14725 [44:50<6:49:10,  1.84s/it]
  9%|▉         | 1387/14725 [44:51<6:48:16,  1.84s/it]
  9%|▉         | 1388/14725 [44:53<6:42:33,  1.81s/it]
  9%|▉         | 1389/14725 [44:55<6:40:29,  1.80s/it]
  9%|▉         | 1390/14725 [44:57<6:39:03,  1.80s/it]
              
+0: {'loss': 0.9013, 'grad_norm': 0.8396625633983624, 'learning_rate': 2e-05, 'epoch': 0.09}
+0: {'loss': 0.925, 'grad_norm': 0.8633832622097125, 'learning_rate': 2e-05, 'epoch': 0.1}
+0:                                         

  9%|▉         | 1390/14725 [44:57<6:39:03,  1.80s/it]
  9%|▉         | 1391/14725 [44:59<6:42:43,  1.81s/it]
  9%|▉         | 1392/14725 [45:00<6:44:12,  1.82s/it]
  9%|▉         | 1393/14725 [45:02<6:45:38,  1.83s/it]
  9%|▉         | 1394/14725 [45:04<6:49:45,  1.84s/it]
  9%|▉         | 1395/14725 [45:06<6:47:03,  1.83s/it]
  9%|▉         | 1396/14725 [45:08<6:50:02,  1.85s/it]
  9%|▉         | 1397/14725 [45:10<6:49:36,  1.84s/it]
  9%|▉         | 1398/14725 [45:12<6:51:34,  1.85s/it]
 10%|▉         | 1399/14725 [45:13<6:51:27,  1.85s/it]
 10%|▉         | 1400/14725 [45:15<6:47:23,  1.83s/it]
                                                      

 10%|▉         | 1400/14725 [45:15<6:47:23,  1.83s/it]
 10%|▉         | 1401/14725 [45:17<6:50:33,  1.85s/it]
 10%|▉         | 1402/14725 [45:19<6:46:26,  1.83s/it]
 10%|▉         | 1403/14725 [45:21<6:50:03,  1.85s/it]
 10%|▉         | 1404/14725 [45:23<6:47:00,  1.83s/it]
 10%|▉      
+0: {'loss': 0.8883, 'grad_norm': 0.9181940892269089, 'learning_rate': 2e-05, 'epoch': 0.1}
+0:    | 1405/14725 [45:24<6:44:05,  1.82s/it]
 10%|▉         | 1406/14725 [45:26<6:43:26,  1.82s/it]
 10%|▉         | 1407/14725 [45:28<6:40:44,  1.81s/it]
 10%|▉         | 1408/14725 [45:30<6:47:54,  1.84s/it]
 10%|▉         | 1409/14725 [45:32<6:43:38,  1.82s/it]
 10%|▉         | 1410/14725 [45:34<6:45:07,  1.83s/it]
                                                      

 10%|▉         | 1410/14725 [45:34<6:45:07,  1.83s/it]
 10%|▉         | 1411/14725 [45:35<6:44:57,  1.82s/it]
 10%|▉         | 1412/14725 [45:37<6:47:03,  1.83s/it]
 10%|▉         | 1413/14725 [45:39<6:48:02,  1.84s/it]
 10%|▉         | 1414/14725 [45:41<6:46:15,  1.83s/it]
 10%|▉         | 1415/14725 [45:43<6:45:10,  1.83s/it]
 10%|▉         | 1416/14725 [45:44<6:42:46,  1.82s/it]
 10%|▉         | 1417/14725 [45:46<6:42:16,  1.81s/it]
 10%|▉         | 1418/14725 [45:48<6:45:00,  1.83s/it]
 10%|▉         | 1419/14725 [45:50<6:46:43,  1.83s/it]
 10%|▉         | 1420/14725 [45:52<6:42:28,  1.82s/it]
             
+0: {'loss': 0.8917, 'grad_norm': 0.8075416219264022, 'learning_rate': 2e-05, 'epoch': 0.1}
+0: {'loss': 0.9011, 'grad_norm': 0.9178806251985487, 'learning_rate': 2e-05, 'epoch': 0.1}
+0:                                          

 10%|▉         | 1420/14725 [45:52<6:42:28,  1.82s/it]
 10%|▉         | 1421/14725 [45:54<6:47:55,  1.84s/it]
 10%|▉         | 1422/14725 [45:55<6:43:57,  1.82s/it]
 10%|▉         | 1423/14725 [45:57<6:41:42,  1.81s/it]
 10%|▉         | 1424/14725 [45:59<6:40:00,  1.80s/it]
 10%|▉         | 1425/14725 [46:01<6:42:02,  1.81s/it]
 10%|▉         | 1426/14725 [46:03<6:38:59,  1.80s/it]
 10%|▉         | 1427/14725 [46:05<6:52:26,  1.86s/it]
 10%|▉         | 1428/14725 [46:06<6:51:14,  1.86s/it]
 10%|▉         | 1429/14725 [46:08<6:50:12,  1.85s/it]
 10%|▉         | 1430/14725 [46:10<6:49:12,  1.85s/it]
                                                      

 10%|▉         | 1430/14725 [46:10<6:49:12,  1.85s/it]
 10%|▉         | 1431/14725 [46:12<6:47:54,  1.84s/it]
 10%|▉         | 1432/14725 [46:14<6:44:26,  1.83s/it]
 10%|▉         | 1433/14725 [46:16<6:41:03,  1.81s/it]
 10%|▉         | 1434/14725 [46:17<6:42:05,  1.82s/it]
 10%|▉     
+0: {'loss': 0.8861, 'grad_norm': 0.8208753573313876, 'learning_rate': 2e-05, 'epoch': 0.1}
+0:     | 1435/14725 [46:19<6:40:35,  1.81s/it]
 10%|▉         | 1436/14725 [46:21<6:42:17,  1.82s/it]
 10%|▉         | 1437/14725 [46:23<6:45:48,  1.83s/it]
 10%|▉         | 1438/14725 [46:25<6:46:30,  1.84s/it]
 10%|▉         | 1439/14725 [46:26<6:43:02,  1.82s/it]
 10%|▉         | 1440/14725 [46:28<6:41:24,  1.81s/it]
                                                      

 10%|▉         | 1440/14725 [46:28<6:41:24,  1.81s/it]
 10%|▉         | 1441/14725 [46:30<6:47:18,  1.84s/it]
 10%|▉         | 1442/14725 [46:32<6:50:55,  1.86s/it]
 10%|▉         | 1443/14725 [46:34<6:51:30,  1.86s/it]
 10%|▉         | 1444/14725 [46:36<6:46:57,  1.84s/it]
 10%|▉         | 1445/14725 [46:38<6:49:49,  1.85s/it]
 10%|▉         | 1446/14725 [46:39<6:47:58,  1.84s/it]
 10%|▉         | 1447/14725 [46:41<6:48:01,  1.84s/it]
 10%|▉         | 1448/14725 [46:43<6:46:54,  1.84s/it]
 10%|▉         | 1449/14725 [46:45<6:45:19,  1.83s/it]
 10%|▉         | 1450/14725 [46:47<6:41:11,  1.81s/it]
            
+0: {'loss': 0.9109, 'grad_norm': 0.9231401199456436, 'learning_rate': 2e-05, 'epoch': 0.1}
+0: {'loss': 0.9195, 'grad_norm': 1.0271270443930942, 'learning_rate': 2e-05, 'epoch': 0.1}
+0:                                           

 10%|▉         | 1450/14725 [46:47<6:41:11,  1.81s/it]
 10%|▉         | 1451/14725 [46:49<6:45:54,  1.83s/it]
 10%|▉         | 1452/14725 [46:50<6:47:31,  1.84s/it]
 10%|▉         | 1453/14725 [46:52<6:48:11,  1.85s/it]
 10%|▉         | 1454/14725 [46:54<6:49:22,  1.85s/it]
 10%|▉         | 1455/14725 [46:56<6:45:52,  1.84s/it]
 10%|▉         | 1456/14725 [46:58<6:41:39,  1.82s/it]
 10%|▉         | 1457/14725 [47:00<6:41:27,  1.82s/it]
 10%|▉         | 1458/14725 [47:01<6:38:20,  1.80s/it]
 10%|▉         | 1459/14725 [47:03<6:41:13,  1.81s/it]
 10%|▉         | 1460/14725 [47:05<6:38:03,  1.80s/it]
                                                      

 10%|▉         | 1460/14725 [47:05<6:38:03,  1.80s/it]
 10%|▉         | 1461/14725 [47:07<6:37:29,  1.80s/it]
 10%|▉         | 1462/14725 [47:09<6:38:12,  1.80s/it]
 10%|▉         | 1463/14725 [47:10<6:40:50,  1.81s/it]
 10%|▉         | 1464/14725 [47:12<6:37:11,  1.80s/it]
 10%|▉    
+0: {'loss': 0.912, 'grad_norm': 0.7974082329913675, 'learning_rate': 2e-05, 'epoch': 0.1}
+0:      | 1465/14725 [47:14<6:38:09,  1.80s/it]
 10%|▉         | 1466/14725 [47:16<6:42:36,  1.82s/it]
 10%|▉         | 1467/14725 [47:18<6:41:01,  1.81s/it]
 10%|▉         | 1468/14725 [47:19<6:37:26,  1.80s/it]
 10%|▉         | 1469/14725 [47:21<6:35:35,  1.79s/it]
 10%|▉         | 1470/14725 [47:23<6:35:14,  1.79s/it]
                                                      

 10%|▉         | 1470/14725 [47:23<6:35:14,  1.79s/it]
 10%|▉         | 1471/14725 [47:25<6:39:00,  1.81s/it]
 10%|▉         | 1472/14725 [47:27<6:42:25,  1.82s/it]
 10%|█         | 1473/14725 [47:28<6:42:56,  1.82s/it]
 10%|█         | 1474/14725 [47:30<6:39:30,  1.81s/it]
 10%|█         | 1475/14725 [47:32<6:40:35,  1.81s/it]
 10%|█         | 1476/14725 [47:34<6:37:43,  1.80s/it]
 10%|█         | 1477/14725 [47:36<6:35:50,  1.79s/it]
 10%|█         | 1478/14725 [47:37<6:42:01,  1.82s/it]
 10%|█         | 1479/14725 [47:39<6:39:57,  1.81s/it]
 10%|█         | 1480/14725 [47:41<6:46:48,  1.84s/it]
           
+0: {'loss': 0.9045, 'grad_norm': 0.856414322733528, 'learning_rate': 2e-05, 'epoch': 0.1}
+0: {'loss': 0.9142, 'grad_norm': 0.984160250978165, 'learning_rate': 2e-05, 'epoch': 0.1}
+0:                                            

 10%|█         | 1480/14725 [47:41<6:46:48,  1.84s/it]
 10%|█         | 1481/14725 [47:43<6:44:52,  1.83s/it]
 10%|█         | 1482/14725 [47:45<6:46:22,  1.84s/it]
 10%|█         | 1483/14725 [47:47<6:51:28,  1.86s/it]
 10%|█         | 1484/14725 [47:49<6:46:13,  1.84s/it]
 10%|█         | 1485/14725 [47:50<6:44:32,  1.83s/it]
 10%|█         | 1486/14725 [47:52<6:43:30,  1.83s/it]
 10%|█         | 1487/14725 [47:54<6:41:57,  1.82s/it]
 10%|█         | 1488/14725 [47:56<6:41:23,  1.82s/it]
 10%|█         | 1489/14725 [47:58<6:40:21,  1.81s/it]
 10%|█         | 1490/14725 [47:59<6:39:04,  1.81s/it]
                                                      

 10%|█         | 1490/14725 [47:59<6:39:04,  1.81s/it]
 10%|█         | 1491/14725 [48:01<6:39:53,  1.81s/it]
 10%|█         | 1492/14725 [48:03<6:46:56,  1.85s/it]
 10%|█         | 1493/14725 [48:05<6:50:58,  1.86s/it]
 10%|█         | 1494/14725 [48:07<6:46:52,  1.85s/it]
 10%|█   
+0: {'loss': 0.9045, 'grad_norm': 1.08317434391019, 'learning_rate': 2e-05, 'epoch': 0.1}
+0:       | 1495/14725 [48:09<6:43:21,  1.83s/it]
 10%|█         | 1496/14725 [48:10<6:39:20,  1.81s/it]
 10%|█         | 1497/14725 [48:12<6:37:38,  1.80s/it]
 10%|█         | 1498/14725 [48:14<6:36:07,  1.80s/it]
 10%|█         | 1499/14725 [48:16<6:34:39,  1.79s/it]
 10%|█         | 1500/14725 [48:18<6:39:28,  1.81s/it]
                                                      

 10%|█         | 1500/14725 [48:18<6:39:28,  1.81s/it]
 10%|█         | 1501/14725 [48:20<6:44:29,  1.84s/it]
 10%|█         | 1502/14725 [48:21<6:46:33,  1.84s/it]
 10%|█         | 1503/14725 [48:23<6:44:32,  1.84s/it]
 10%|█         | 1504/14725 [48:25<6:44:43,  1.84s/it]
 10%|█         | 1505/14725 [48:27<6:44:30,  1.84s/it]
 10%|█         | 1506/14725 [48:29<6:39:46,  1.81s/it]
 10%|█         | 1507/14725 [48:30<6:39:32,  1.81s/it]
 10%|█         | 1508/14725 [48:32<6:40:39,  1.82s/it]
 10%|█         | 1509/14725 [48:34<6:40:25,  1.82s/it]
 10%|█         | 1510/14725 [48:36<6:39:50,  1.82s/it]
          
+0: {'loss': 0.9039, 'grad_norm': 0.8439976238218464, 'learning_rate': 2e-05, 'epoch': 0.1}
+0: {'loss': 0.9032, 'grad_norm': 0.9355639642344662, 'learning_rate': 2e-05, 'epoch': 0.1}
+0:                                             

 10%|█         | 1510/14725 [48:36<6:39:50,  1.82s/it]
 10%|█         | 1511/14725 [48:38<6:42:20,  1.83s/it]
 10%|█         | 1512/14725 [48:40<6:41:30,  1.82s/it]
 10%|█         | 1513/14725 [48:41<6:45:55,  1.84s/it]
 10%|█         | 1514/14725 [48:43<6:42:48,  1.83s/it]
 10%|█         | 1515/14725 [48:45<6:41:46,  1.82s/it]
 10%|█         | 1516/14725 [48:47<6:42:07,  1.83s/it]
 10%|█         | 1517/14725 [48:49<6:42:59,  1.83s/it]
 10%|█         | 1518/14725 [48:51<6:41:17,  1.82s/it]
 10%|█         | 1519/14725 [48:52<6:39:18,  1.81s/it]
 10%|█         | 1520/14725 [48:54<6:42:59,  1.83s/it]
                                                      

 10%|█         | 1520/14725 [48:54<6:42:59,  1.83s/it]
 10%|█         | 1521/14725 [48:56<6:39:59,  1.82s/it]
 10%|█         | 1522/14725 [48:58<6:40:41,  1.82s/it]
 10%|█         | 1523/14725 [49:00<6:38:24,  1.81s/it]
 10%|█         | 1524/14725 [49:01<6:40:15,  1.82s/it]
 10%|█  
+0: {'loss': 0.8816, 'grad_norm': 0.9158062192038211, 'learning_rate': 2e-05, 'epoch': 0.1}
+0:        | 1525/14725 [49:03<6:37:00,  1.80s/it]
 10%|█         | 1526/14725 [49:05<6:35:34,  1.80s/it]
 10%|█         | 1527/14725 [49:07<6:36:25,  1.80s/it]
 10%|█         | 1528/14725 [49:09<6:35:23,  1.80s/it]
 10%|█         | 1529/14725 [49:10<6:35:45,  1.80s/it]
 10%|█         | 1530/14725 [49:12<6:35:19,  1.80s/it]
                                                      

 10%|█         | 1530/14725 [49:12<6:35:19,  1.80s/it]
 10%|█         | 1531/14725 [49:14<6:34:02,  1.79s/it]
 10%|█         | 1532/14725 [49:16<6:34:47,  1.80s/it]
 10%|█         | 1533/14725 [49:18<6:36:15,  1.80s/it]
 10%|█         | 1534/14725 [49:20<6:42:14,  1.83s/it]
 10%|█         | 1535/14725 [49:21<6:38:58,  1.81s/it]
 10%|█         | 1536/14725 [49:23<6:41:52,  1.83s/it]
 10%|█         | 1537/14725 [49:25<6:39:39,  1.82s/it]
 10%|█         | 1538/14725 [49:27<6:40:01,  1.82s/it]
 10%|█         | 1539/14725 [49:29<6:38:14,  1.81s/it]
 10%|█         | 1540/14725 [49:30<6:35:38,  1.80s/it]
         
+0: {'loss': 0.8759, 'grad_norm': 0.8368951157446296, 'learning_rate': 2e-05, 'epoch': 0.1}
+0: {'loss': 0.9007, 'grad_norm': 0.9177433754840092, 'learning_rate': 2e-05, 'epoch': 0.11}
+0:                                              

 10%|█         | 1540/14725 [49:30<6:35:38,  1.80s/it]
 10%|█         | 1541/14725 [49:32<6:38:54,  1.82s/it]
 10%|█         | 1542/14725 [49:34<6:37:51,  1.81s/it]
 10%|█         | 1543/14725 [49:36<6:38:41,  1.81s/it]
 10%|█         | 1544/14725 [49:38<6:41:52,  1.83s/it]
 10%|█         | 1545/14725 [49:39<6:39:22,  1.82s/it]
 10%|█         | 1546/14725 [49:41<6:40:19,  1.82s/it]
 11%|█         | 1547/14725 [49:43<6:38:52,  1.82s/it]
 11%|█         | 1548/14725 [49:45<6:36:39,  1.81s/it]
 11%|█         | 1549/14725 [49:47<6:39:24,  1.82s/it]
 11%|█         | 1550/14725 [49:49<6:41:53,  1.83s/it]
                                                      

 11%|█         | 1550/14725 [49:49<6:41:53,  1.83s/it]
 11%|█         | 1551/14725 [49:50<6:44:17,  1.84s/it]
 11%|█         | 1552/14725 [49:52<6:45:51,  1.85s/it]
 11%|█         | 1553/14725 [49:54<6:43:26,  1.84s/it]
 11%|█         | 1554/14725 [49:56<6:54:55,  1.89s/it]
 11%|█ 
+0: {'loss': 0.9088, 'grad_norm': 1.0719316302978696, 'learning_rate': 2e-05, 'epoch': 0.11}
+0:         | 1555/14725 [49:58<6:48:59,  1.86s/it]
 11%|█         | 1556/14725 [50:00<6:49:59,  1.87s/it]
 11%|█         | 1557/14725 [50:02<6:51:05,  1.87s/it]
 11%|█         | 1558/14725 [50:04<6:46:14,  1.85s/it]
 11%|█         | 1559/14725 [50:05<6:48:30,  1.86s/it]
 11%|█         | 1560/14725 [50:07<6:50:23,  1.87s/it]
                                                      

 11%|█         | 1560/14725 [50:07<6:50:23,  1.87s/it]
 11%|█         | 1561/14725 [50:09<6:49:37,  1.87s/it]
 11%|█         | 1562/14725 [50:11<6:47:38,  1.86s/it]
 11%|█         | 1563/14725 [50:13<6:48:10,  1.86s/it]
 11%|█         | 1564/14725 [50:15<6:44:52,  1.85s/it]
 11%|█         | 1565/14725 [50:17<6:48:38,  1.86s/it]
 11%|█         | 1566/14725 [50:18<6:48:04,  1.86s/it]
 11%|█         | 1567/14725 [50:20<6:51:00,  1.87s/it]
 11%|█         | 1568/14725 [50:22<6:49:17,  1.87s/it]
 11%|█         | 1569/14725 [50:24<6:43:45,  1.84s/it]
 11%|█         | 1570/14725 [50:26<6:43:05,  1.84s/it]
        
+0: {'loss': 0.9108, 'grad_norm': 0.869362927510703, 'learning_rate': 2e-05, 'epoch': 0.11}
+0: {'loss': 0.8832, 'grad_norm': 0.9900073165811678, 'learning_rate': 2e-05, 'epoch': 0.11}
+0:                                               

 11%|█         | 1570/14725 [50:26<6:43:05,  1.84s/it]
 11%|█         | 1571/14725 [50:28<6:48:38,  1.86s/it]
 11%|█         | 1572/14725 [50:30<6:44:58,  1.85s/it]
 11%|█         | 1573/14725 [50:31<6:43:04,  1.84s/it]
 11%|█         | 1574/14725 [50:33<6:40:39,  1.83s/it]
 11%|█         | 1575/14725 [50:35<6:38:02,  1.82s/it]
 11%|█         | 1576/14725 [50:37<6:37:24,  1.81s/it]
 11%|█         | 1577/14725 [50:39<6:35:04,  1.80s/it]
 11%|█         | 1578/14725 [50:40<6:40:48,  1.83s/it]
 11%|█         | 1579/14725 [50:42<6:41:45,  1.83s/it]
 11%|█         | 1580/14725 [50:44<6:40:34,  1.83s/it]
                                                      

 11%|█         | 1580/14725 [50:44<6:40:34,  1.83s/it]
 11%|█         | 1581/14725 [50:46<6:40:00,  1.83s/it]
 11%|█         | 1582/14725 [50:48<6:44:37,  1.85s/it]
 11%|█         | 1583/14725 [50:50<6:45:08,  1.85s/it]
 11%|█         | 1584/14725 [50:52<6:54:04,  1.89s/it]
 11%|█
+0: {'loss': 0.8764, 'grad_norm': 0.7934519614870558, 'learning_rate': 2e-05, 'epoch': 0.11}
+0:          | 1585/14725 [50:53<6:47:02,  1.86s/it]
 11%|█         | 1586/14725 [50:55<6:42:39,  1.84s/it]
 11%|█         | 1587/14725 [50:57<6:41:20,  1.83s/it]
 11%|█         | 1588/14725 [50:59<6:47:52,  1.86s/it]
 11%|█         | 1589/14725 [51:01<6:40:25,  1.83s/it]
 11%|█         | 1590/14725 [51:03<6:39:05,  1.82s/it]
                                                      

 11%|█         | 1590/14725 [51:03<6:39:05,  1.82s/it]
 11%|█         | 1591/14725 [51:04<6:38:33,  1.82s/it]
 11%|█         | 1592/14725 [51:06<6:41:14,  1.83s/it]
 11%|█         | 1593/14725 [51:08<6:40:32,  1.83s/it]
 11%|█         | 1594/14725 [51:10<6:39:22,  1.82s/it]
 11%|█         | 1595/14725 [51:12<6:35:24,  1.81s/it]
 11%|█         | 1596/14725 [51:13<6:32:07,  1.79s/it]
 11%|█         | 1597/14725 [51:15<6:31:36,  1.79s/it]
 11%|█         | 1598/14725 [51:17<6:35:37,  1.81s/it]
 11%|█         | 1599/14725 [51:19<6:33:42,  1.80s/it]
 11%|█         | 1600/14725 [51:21<6:38:58,  1.82s/it]
       
+0: {'loss': 0.8898, 'grad_norm': 0.8892482398257261, 'learning_rate': 2e-05, 'epoch': 0.11}
+0: {'loss': 0.9079, 'grad_norm': 0.8624363449552919, 'learning_rate': 2e-05, 'epoch': 0.11}
+0:                                                

 11%|█         | 1600/14725 [51:21<6:38:58,  1.82s/it]
 11%|█         | 1601/14725 [51:22<6:37:17,  1.82s/it]
 11%|█         | 1602/14725 [51:24<6:38:09,  1.82s/it]
 11%|█         | 1603/14725 [51:26<6:38:21,  1.82s/it]
 11%|█         | 1604/14725 [51:28<6:36:20,  1.81s/it]
 11%|█         | 1605/14725 [51:30<6:40:35,  1.83s/it]
 11%|█         | 1606/14725 [51:32<6:36:20,  1.81s/it]
 11%|█         | 1607/14725 [51:33<6:38:35,  1.82s/it]
 11%|█         | 1608/14725 [51:35<6:42:56,  1.84s/it]
 11%|█         | 1609/14725 [51:37<6:39:14,  1.83s/it]
 11%|█         | 1610/14725 [51:39<6:37:43,  1.82s/it]
                                                      

 11%|█         | 1610/14725 [51:39<6:37:43,  1.82s/it]
 11%|█         | 1611/14725 [51:41<6:38:27,  1.82s/it]
 11%|█         | 1612/14725 [51:42<6:35:17,  1.81s/it]
 11%|█         | 1613/14725 [51:44<6:35:01,  1.81s/it]
 11%|█         | 1614/14725 [51:46<6:34:26,  1.81s/it]
 11%|�
+0: {'loss': 0.9171, 'grad_norm': 0.8129968391651631, 'learning_rate': 2e-05, 'epoch': 0.11}
+0: �         | 1615/14725 [51:48<6:35:03,  1.81s/it]
 11%|█         | 1616/14725 [51:50<6:35:41,  1.81s/it]
 11%|█         | 1617/14725 [51:52<6:40:49,  1.83s/it]
 11%|█         | 1618/14725 [51:53<6:38:21,  1.82s/it]
 11%|█         | 1619/14725 [51:55<6:35:41,  1.81s/it]
 11%|█         | 1620/14725 [51:57<6:34:39,  1.81s/it]
                                                      

 11%|█         | 1620/14725 [51:57<6:34:39,  1.81s/it]
 11%|█         | 1621/14725 [51:59<6:33:50,  1.80s/it]
 11%|█         | 1622/14725 [52:01<6:35:05,  1.81s/it]
 11%|█         | 1623/14725 [52:02<6:33:56,  1.80s/it]
 11%|█         | 1624/14725 [52:04<6:32:48,  1.80s/it]
 11%|█         | 1625/14725 [52:06<6:34:20,  1.81s/it]
 11%|█         | 1626/14725 [52:08<6:33:44,  1.80s/it]
 11%|█         | 1627/14725 [52:10<6:32:50,  1.80s/it]
 11%|█         | 1628/14725 [52:11<6:33:39,  1.80s/it]
 11%|█         | 1629/14725 [52:13<6:31:56,  1.80s/it]
 11%|█         | 1630/14725 [52:15<6:31:07,  1.79s/it]
      
+0: {'loss': 0.8823, 'grad_norm': 0.7951782029736295, 'learning_rate': 2e-05, 'epoch': 0.11}
+0: {'loss': 0.8795, 'grad_norm': 0.8464082196841677, 'learning_rate': 2e-05, 'epoch': 0.11}
+0:                                                 

 11%|█         | 1630/14725 [52:15<6:31:07,  1.79s/it]
 11%|█         | 1631/14725 [52:17<6:35:55,  1.81s/it]
 11%|█         | 1632/14725 [52:19<6:35:44,  1.81s/it]
 11%|█         | 1633/14725 [52:20<6:38:54,  1.83s/it]
 11%|█         | 1634/14725 [52:22<6:42:41,  1.85s/it]
 11%|█         | 1635/14725 [52:24<6:40:03,  1.83s/it]
 11%|█         | 1636/14725 [52:26<6:37:51,  1.82s/it]
 11%|█         | 1637/14725 [52:28<6:35:43,  1.81s/it]
 11%|█         | 1638/14725 [52:30<6:40:03,  1.83s/it]
 11%|█         | 1639/14725 [52:31<6:39:58,  1.83s/it]
 11%|█         | 1640/14725 [52:33<6:44:13,  1.85s/it]
                                                      

 11%|█         | 1640/14725 [52:33<6:44:13,  1.85s/it]
 11%|█         | 1641/14725 [52:35<6:39:52,  1.83s/it]
 11%|█         | 1642/14725 [52:37<6:43:31,  1.85s/it]
 11%|█         | 1643/14725 [52:39<6:41:45,  1.84s/it]
 11%|█         | 1644/14725 [52:41<6:37:51,  1.82s/it]
 11%|�
+0: {'loss': 0.8719, 'grad_norm': 0.8926714805590898, 'learning_rate': 2e-05, 'epoch': 0.11}
+0: ��         | 1645/14725 [52:42<6:35:38,  1.81s/it]
 11%|█         | 1646/14725 [52:44<6:38:41,  1.83s/it]
 11%|█         | 1647/14725 [52:46<6:55:26,  1.91s/it]
 11%|█         | 1648/14725 [52:48<6:49:54,  1.88s/it]
 11%|█         | 1649/14725 [52:50<6:51:25,  1.89s/it]
 11%|█         | 1650/14725 [52:52<6:45:08,  1.86s/it]
                                                      

 11%|█         | 1650/14725 [52:52<6:45:08,  1.86s/it]
 11%|█         | 1651/14725 [52:54<6:41:30,  1.84s/it]
 11%|█         | 1652/14725 [52:56<6:38:43,  1.83s/it]
 11%|█         | 1653/14725 [52:57<6:35:26,  1.82s/it]
 11%|█         | 1654/14725 [52:59<6:39:40,  1.83s/it]
 11%|█         | 1655/14725 [53:01<6:35:22,  1.82s/it]
 11%|█         | 1656/14725 [53:03<6:34:14,  1.81s/it]
 11%|█▏        | 1657/14725 [53:05<6:33:15,  1.81s/it]
 11%|█▏        | 1658/14725 [53:06<6:37:57,  1.83s/it]
 11%|█▏        | 1659/14725 [53:08<6:39:19,  1.83s/it]
 11%|█▏        | 1660/14725 [53:10<6:37:38,  1.83s/i
+0: {'loss': 0.8982, 'grad_norm': 0.8486982718564461, 'learning_rate': 2e-05, 'epoch': 0.11}
+0: {'loss': 0.8973, 'grad_norm': 0.9343121842459932, 'learning_rate': 2e-05, 'epoch': 0.11}
+0: t]
                                                      

 11%|█▏        | 1660/14725 [53:10<6:37:38,  1.83s/it]
 11%|█▏        | 1661/14725 [53:12<6:37:41,  1.83s/it]
 11%|█▏        | 1662/14725 [53:14<6:33:27,  1.81s/it]
 11%|█▏        | 1663/14725 [53:15<6:31:18,  1.80s/it]
 11%|█▏        | 1664/14725 [53:17<6:31:09,  1.80s/it]
 11%|█▏        | 1665/14725 [53:19<6:34:00,  1.81s/it]
 11%|█▏        | 1666/14725 [53:21<6:33:33,  1.81s/it]
 11%|█▏        | 1667/14725 [53:23<6:36:18,  1.82s/it]
 11%|█▏        | 1668/14725 [53:25<6:37:40,  1.83s/it]
 11%|█▏        | 1669/14725 [53:26<6:36:14,  1.82s/it]
 11%|█▏        | 1670/14725 [53:28<6:37:42,  1.83s/it]
                                                      

 11%|█▏        | 1670/14725 [53:28<6:37:42,  1.83s/it]
 11%|█▏        | 1671/14725 [53:30<6:39:00,  1.83s/it]
 11%|█▏        | 1672/14725 [53:32<6:34:06,  1.81s/it]
 11%|█▏        | 1673/14725 [53:34<6:32:20,  1.80s/it]
 11%|█▏        | 167
+0: {'loss': 0.9048, 'grad_norm': 0.8020511452355631, 'learning_rate': 2e-05, 'epoch': 0.11}
+0: 4/14725 [53:35<6:33:33,  1.81s/it]
 11%|█▏        | 1675/14725 [53:37<6:35:57,  1.82s/it]
 11%|█▏        | 1676/14725 [53:39<6:36:15,  1.82s/it]
 11%|█▏        | 1677/14725 [53:41<6:34:44,  1.82s/it]
 11%|█▏        | 1678/14725 [53:43<6:33:00,  1.81s/it]
 11%|█▏        | 1679/14725 [53:45<6:53:11,  1.90s/it]
 11%|█▏        | 1680/14725 [53:47<6:44:23,  1.86s/it]
                                                      

 11%|█▏        | 1680/14725 [53:47<6:44:23,  1.86s/it]
 11%|█▏        | 1681/14725 [53:48<6:43:34,  1.86s/it]
 11%|█▏        | 1682/14725 [53:50<6:40:49,  1.84s/it]
 11%|█▏        | 1683/14725 [53:52<6:40:06,  1.84s/it]
 11%|█▏        | 1684/14725 [53:54<6:41:55,  1.85s/it]
 11%|█▏        | 1685/14725 [53:56<6:38:40,  1.83s/it]
 11%|█▏        | 1686/14725 [53:58<6:39:18,  1.84s/it]
 11%|█▏        | 1687/14725 [53:59<6:38:47,  1.84s/it]
 11%|█▏        | 1688/14725 [54:01<6:38:07,  1.83s/it]
 11%|█▏        | 1689/14725 [54:03<6:38:30, 
+0: {'loss': 0.9068, 'grad_norm': 0.7936481460271637, 'learning_rate': 2e-05, 'epoch': 0.11}
+0: {'loss': 0.9036, 'grad_norm': 0.8397208905667551, 'learning_rate': 2e-05, 'epoch': 0.12}
+0:  1.83s/it]
 11%|█▏        | 1690/14725 [54:05<6:38:22,  1.83s/it]
                                                      

 11%|█▏        | 1690/14725 [54:05<6:38:22,  1.83s/it]
 11%|█▏        | 1691/14725 [54:07<6:39:11,  1.84s/it]
 11%|█▏        | 1692/14725 [54:09<6:41:32,  1.85s/it]
 11%|█▏        | 1693/14725 [54:11<6:41:34,  1.85s/it]
 12%|█▏        | 1694/14725 [54:12<6:39:11,  1.84s/it]
 12%|█▏        | 1695/14725 [54:15<7:05:17,  1.96s/it]
 12%|█▏        | 1696/14725 [54:16<6:53:21,  1.90s/it]
 12%|█▏        | 1697/14725 [54:18<7:08:01,  1.97s/it]
 12%|█▏        | 1698/14725 [54:20<7:00:04,  1.93s/it]
 12%|█▏        | 1699/14725 [54:22<6:54:39,  1.91s/it]
 12%|█▏        | 1700/14725 [54:24<6:53:29,  1.90s/it]
                                                      

 12%|█▏        | 1700/14725 [54:24<6:53:29,  1.90s/it]
 12%|█▏        | 1701/14725 [54:26<7:09:43,  1.98s/it]
 12%|█▏        | 1702/14725 [54:28<6:59:10,  1.93s/it]
 12%|█▏     
+0: {'loss': 0.8896, 'grad_norm': 0.8764779432683711, 'learning_rate': 2e-05, 'epoch': 0.12}
+0:    | 1703/14725 [54:30<6:53:19,  1.90s/it]
 12%|█▏        | 1704/14725 [54:32<6:44:30,  1.86s/it]
 12%|█▏        | 1705/14725 [54:34<6:46:08,  1.87s/it]
 12%|█▏        | 1706/14725 [54:35<6:43:08,  1.86s/it]
 12%|█▏        | 1707/14725 [54:37<6:38:28,  1.84s/it]
 12%|█▏        | 1708/14725 [54:39<6:39:20,  1.84s/it]
 12%|█▏        | 1709/14725 [54:41<6:38:00,  1.83s/it]
 12%|█▏        | 1710/14725 [54:43<6:38:27,  1.84s/it]
                                                      

 12%|█▏        | 1710/14725 [54:43<6:38:27,  1.84s/it]
 12%|█▏        | 1711/14725 [54:45<6:39:39,  1.84s/it]
 12%|█▏        | 1712/14725 [54:46<6:40:27,  1.85s/it]
 12%|█▏        | 1713/14725 [54:48<6:41:51,  1.85s/it]
 12%|█▏        | 1714/14725 [54:50<6:51:17,  1.90s/it]
 12%|█▏        | 1715/14725 [54:52<6:54:09,  1.91s/it]
 12%|█▏        | 1716/14725 [54:54<6:48:50,  1.89s/it]
 12%|█▏        | 1717/14725 [54:56<6:42:11,  1.86s/it]
 12%|█▏        | 1718/14725 [54:58<6
+0: {'loss': 0.9196, 'grad_norm': 0.9719053048821865, 'learning_rate': 2e-05, 'epoch': 0.12}
+0: {'loss': 0.8868, 'grad_norm': 0.8477620162171536, 'learning_rate': 2e-05, 'epoch': 0.12}
+0: :38:40,  1.84s/it]
 12%|█▏        | 1719/14725 [54:59<6:36:42,  1.83s/it]
 12%|█▏        | 1720/14725 [55:01<6:32:24,  1.81s/it]
                                                      

 12%|█▏        | 1720/14725 [55:01<6:32:24,  1.81s/it]
 12%|█▏        | 1721/14725 [55:03<6:35:44,  1.83s/it]
 12%|█▏        | 1722/14725 [55:05<6:31:37,  1.81s/it]
 12%|█▏        | 1723/14725 [55:07<6:34:47,  1.82s/it]
 12%|█▏        | 1724/14725 [55:08<6:35:55,  1.83s/it]
 12%|█▏        | 1725/14725 [55:10<6:31:31,  1.81s/it]
 12%|█▏        | 1726/14725 [55:12<6:31:34,  1.81s/it]
 12%|█▏        | 1727/14725 [55:14<6:34:16,  1.82s/it]
 12%|█▏        | 1728/14725 [55:16<6:51:49,  1.90s/it]
 12%|█▏        | 1729/14725 [55:18<6:47:03,  1.88s/it]
 12%|█▏        | 1730/14725 [55:20<6:41:51,  1.86s/it]
                                                      

 12%|█▏        | 1730/14725 [55:20<6:41:51,  1.86s/it]
 12%|█▏        | 1731/14725 [55:21<6:37:30,  1.84s/it]
 12%|█
+0: {'loss': 0.8866, 'grad_norm': 0.9143646826761619, 'learning_rate': 2e-05, 'epoch': 0.12}
+0: ▏        | 1732/14725 [55:23<6:42:39,  1.86s/it]
 12%|█▏        | 1733/14725 [55:25<6:40:38,  1.85s/it]
 12%|█▏        | 1734/14725 [55:27<6:42:14,  1.86s/it]
 12%|█▏        | 1735/14725 [55:29<6:42:09,  1.86s/it]
 12%|█▏        | 1736/14725 [55:31<6:41:39,  1.86s/it]
 12%|█▏        | 1737/14725 [55:33<6:41:20,  1.85s/it]
 12%|█▏        | 1738/14725 [55:34<6:39:14,  1.84s/it]
 12%|█▏        | 1739/14725 [55:36<6:34:29,  1.82s/it]
 12%|█▏        | 1740/14725 [55:38<6:33:19,  1.82s/it]
                                                      

 12%|█▏        | 1740/14725 [55:38<6:33:19,  1.82s/it]
 12%|█▏        | 1741/14725 [55:40<6:36:59,  1.83s/it]
 12%|█▏        | 1742/14725 [55:42<6:33:45,  1.82s/it]
 12%|█▏        | 1743/14725 [55:43<6:30:47,  1.81s/it]
 12%|█▏        | 1744/14725 [55:45<6:32:03,  1.81s/it]
 12%|█▏        | 1745/14725 [55:47<6:32:29,  1.81s/it]
 12%|█▏        | 1746/14725 [55:49<6:35:41,  1.83s/it]
 12%|█▏        | 1747/14725 
+0: {'loss': 0.8935, 'grad_norm': 0.9329298247934478, 'learning_rate': 2e-05, 'epoch': 0.12}
+0: {'loss': 0.9058, 'grad_norm': 0.9213373723268851, 'learning_rate': 2e-05, 'epoch': 0.12}
+0: [55:51<6:34:15,  1.82s/it]
 12%|█▏        | 1748/14725 [55:53<6:31:07,  1.81s/it]
 12%|█▏        | 1749/14725 [55:54<6:34:29,  1.82s/it]
 12%|█▏        | 1750/14725 [55:56<6:33:53,  1.82s/it]
                                                      

 12%|█▏        | 1750/14725 [55:56<6:33:53,  1.82s/it]
 12%|█▏        | 1751/14725 [55:58<6:33:23,  1.82s/it]
 12%|█▏        | 1752/14725 [56:00<6:34:09,  1.82s/it]
 12%|█▏        | 1753/14725 [56:02<6:32:16,  1.81s/it]
 12%|█▏        | 1754/14725 [56:03<6:29:51,  1.80s/it]
 12%|█▏        | 1755/14725 [56:05<6:34:38,  1.83s/it]
 12%|█▏        | 1756/14725 [56:07<6:32:25,  1.82s/it]
 12%|█▏        | 1757/14725 [56:09<6:31:36,  1.81s/it]
 12%|█▏        | 1758/14725 [56:11<6:30:05,  1.81s/it]
 12%|█▏        | 1759/14725 [56:13<6:32:43,  1.82s/it]
 12%|█▏        | 1760/14725 [56:14<6:40:00,  1.85s/it]
                                                      

 12%|█▏        | 1760/14725 [56:14<6:40:00,  1.85s/it]
+0: {'loss': 0.8953, 'grad_norm': 0.8057426885444958, 'learning_rate': 2e-05, 'epoch': 0.12}
+0:  12%|█▏        | 1761/14725 [56:16<6:36:31,  1.84s/it]
 12%|█▏        | 1762/14725 [56:18<6:39:15,  1.85s/it]
 12%|█▏        | 1763/14725 [56:20<6:38:04,  1.84s/it]
 12%|█▏        | 1764/14725 [56:22<6:35:11,  1.83s/it]
 12%|█▏        | 1765/14725 [56:24<6:36:53,  1.84s/it]
 12%|█▏        | 1766/14725 [56:25<6:36:12,  1.83s/it]
 12%|█▏        | 1767/14725 [56:27<6:39:43,  1.85s/it]
 12%|█▏        | 1768/14725 [56:29<6:35:14,  1.83s/it]
 12%|█▏        | 1769/14725 [56:31<6:34:28,  1.83s/it]
 12%|█▏        | 1770/14725 [56:33<6:35:21,  1.83s/it]
                                                      

 12%|█▏        | 1770/14725 [56:33<6:35:21,  1.83s/it]
 12%|█▏        | 1771/14725 [56:35<6:33:59,  1.82s/it]
 12%|█▏        | 1772/14725 [56:36<6:35:55,  1.83s/it]
 12%|█▏        | 1773/14725 [56:38<6:31:36,  1.81s/it]
 12%|█▏        | 1774/14725 [56:40<6:31:24,  1.81s/it]
 12%|█▏        | 1775/14725 [56:42<6:30:25,  1.81s/it]
 12%|█▏        | 177
+0: {'loss': 0.9109, 'grad_norm': 0.832076001947435, 'learning_rate': 2e-05, 'epoch': 0.12}
+0: 6/14725 [56:44<6:34:21,  1.83s/it]
 12%|█▏        | 1777/14725 [56:46<6:36:03,  1.84s/it]
 12%|█▏        | 1778/14725 [56:47<6:34:29,  1.83s/it]
 12%|█▏        | 1779/14725 [56:49<6:33:15,  1.82s/it]
 12%|█▏        | 1780/14725 [56:51<6:29:07,  1.80s/it]
                                                      

 12%|█▏        | 1780/14725 [56:51<6:29:07,  1.80s/it]
 12%|█▏        | 1781/14725 [56:53<6:27:46,  1.80s/it]
 12%|█▏        | 1782/14725 [56:55<6:31:04,  1.81s/it]
 12%|█▏        | 1783/14725 [56:56<6:31:29,  1.81s/it]
 12%|█▏        | 1784/14725 [56:58<6:31:59,  1.82s/it]
 12%|█▏        | 1785/14725 [57:00<6:29:12,  1.80s/it]
 12%|█▏        | 1786/14725 [57:02<6:33:37,  1.83s/it]
 12%|█▏        | 1787/14725 [57:04<6:31:45,  1.82s/it]
 12%|█▏        | 1788/14725 [57:05<6:34:04,  1.83s/it]
 12%|█▏        | 1789/14725 [57:07<6:34:55,  1.83s/it]
 12%|█▏        | 1790/14725 [57:09<6:31:59,  1.82s/it]
                                                
+0: {'loss': 0.9122, 'grad_norm': 0.8626106668197744, 'learning_rate': 2e-05, 'epoch': 0.12}
+0: {'loss': 0.8769, 'grad_norm': 0.8739967343420733, 'learning_rate': 2e-05, 'epoch': 0.12}
+0:       

 12%|█▏        | 1790/14725 [57:09<6:31:59,  1.82s/it]
 12%|█▏        | 1791/14725 [57:11<6:34:17,  1.83s/it]
 12%|█▏        | 1792/14725 [57:13<6:33:21,  1.82s/it]
 12%|█▏        | 1793/14725 [57:15<6:31:26,  1.82s/it]
 12%|█▏        | 1794/14725 [57:16<6:29:35,  1.81s/it]
 12%|█▏        | 1795/14725 [57:18<6:31:07,  1.81s/it]
 12%|█▏        | 1796/14725 [57:20<6:30:12,  1.81s/it]
 12%|█▏        | 1797/14725 [57:22<6:32:27,  1.82s/it]
 12%|█▏        | 1798/14725 [57:24<6:32:09,  1.82s/it]
 12%|█▏        | 1799/14725 [57:25<6:28:59,  1.81s/it]
 12%|█▏        | 1800/14725 [57:27<6:33:03,  1.82s/it]
                                                      

 12%|█▏        | 1800/14725 [57:27<6:33:03,  1.82s/it]
 12%|█▏        | 1801/14725 [57:29<6:34:06,  1.83s/it]
 12%|█▏        | 1802/14725 [57:31<6:32:14,  1.82s/it]
 12%|█▏        | 1803/14725 [57:33<6:29:19,  1.81s/it]
 12%|█▏        | 1804/14725 [57:35<6:40:34,  1.86s/it]
 12%|█▏     
+0: {'loss': 0.9071, 'grad_norm': 0.8892922123337007, 'learning_rate': 2e-05, 'epoch': 0.12}
+0:    | 1805/14725 [57:37<6:42:22,  1.87s/it]
 12%|█▏        | 1806/14725 [57:38<6:36:50,  1.84s/it]
 12%|█▏        | 1807/14725 [57:40<6:31:50,  1.82s/it]
 12%|█▏        | 1808/14725 [57:42<6:33:24,  1.83s/it]
 12%|█▏        | 1809/14725 [57:44<6:33:36,  1.83s/it]
 12%|█▏        | 1810/14725 [57:46<6:31:21,  1.82s/it]
                                                      

 12%|█▏        | 1810/14725 [57:46<6:31:21,  1.82s/it]
 12%|█▏        | 1811/14725 [57:47<6:33:04,  1.83s/it]
 12%|█▏        | 1812/14725 [57:49<6:31:54,  1.82s/it]
 12%|█▏        | 1813/14725 [57:51<6:37:45,  1.85s/it]
 12%|█▏        | 1814/14725 [57:53<6:33:54,  1.83s/it]
 12%|█▏        | 1815/14725 [57:55<6:30:11,  1.81s/it]
 12%|█▏        | 1816/14725 [57:57<6:30:33,  1.82s/it]
 12%|█▏        | 1817/14725 [57:58<6:33:15,  1.83s/it]
 12%|█▏        | 1818/14725 [58:00<6:36:23,  1.84s/it]
 12%|█▏        | 1819/14725 [58:02<6:32:41,  1.83s/it]
 12%|█▏        | 1820/14725 [58:04<6
+0: {'loss': 0.8909, 'grad_norm': 0.8025480213111036, 'learning_rate': 2e-05, 'epoch': 0.12}
+0: {'loss': 0.9068, 'grad_norm': 0.7883300394507761, 'learning_rate': 2e-05, 'epoch': 0.12}
+0: :33:32,  1.83s/it]
                                                      

 12%|█▏        | 1820/14725 [58:04<6:33:32,  1.83s/it]
 12%|█▏        | 1821/14725 [58:06<6:31:29,  1.82s/it]
 12%|█▏        | 1822/14725 [58:08<6:29:15,  1.81s/it]
 12%|█▏        | 1823/14725 [58:09<6:27:58,  1.80s/it]
 12%|█▏        | 1824/14725 [58:11<6:29:22,  1.81s/it]
 12%|█▏        | 1825/14725 [58:13<6:26:17,  1.80s/it]
 12%|█▏        | 1826/14725 [58:15<6:26:42,  1.80s/it]
 12%|█▏        | 1827/14725 [58:16<6:25:10,  1.79s/it]
 12%|█▏        | 1828/14725 [58:18<6:30:22,  1.82s/it]
 12%|█▏        | 1829/14725 [58:20<6:29:46,  1.81s/it]
 12%|█▏        | 1830/14725 [58:22<6:27:29,  1.80s/it]
                                                      

 12%|█▏        | 1830/14725 [58:22<6:27:29,  1.80s/it]
 12%|█▏        | 1831/14725 [58:24<6:29:03,  1.81s/it]
 12%|█▏        | 1832/14725 [58:26<6:25:55,  1.80s/it]
 12%|█▏        | 1833/14725 [58:27<6:33:14,  1.83s/it]
 12%|��
+0: {'loss': 0.8721, 'grad_norm': 0.8138961963960615, 'learning_rate': 2e-05, 'epoch': 0.12}
+0: ▏        | 1834/14725 [58:29<6:35:49,  1.84s/it]
 12%|█▏        | 1835/14725 [58:31<6:35:16,  1.84s/it]
 12%|█▏        | 1836/14725 [58:33<6:37:14,  1.85s/it]
 12%|█▏        | 1837/14725 [58:35<6:39:07,  1.86s/it]
 12%|█▏        | 1838/14725 [58:37<6:37:24,  1.85s/it]
 12%|█▏        | 1839/14725 [58:39<6:33:55,  1.83s/it]
 12%|█▏        | 1840/14725 [58:40<6:31:05,  1.82s/it]
                                                      

 12%|█▏        | 1840/14725 [58:40<6:31:05,  1.82s/it]
 13%|█▎        | 1841/14725 [58:42<6:31:59,  1.83s/it]
 13%|█▎        | 1842/14725 [58:44<6:29:37,  1.81s/it]
 13%|█▎        | 1843/14725 [58:46<6:27:53,  1.81s/it]
 13%|█▎        | 1844/14725 [58:47<6:26:02,  1.80s/it]
 13%|█▎        | 1845/14725 [58:49<6:31:53,  1.83s/it]
 13%|█▎        | 1846/14725 [58:51<6:33:01,  1.83s/it]
 13%|█▎        | 1847/14725 [58:53<6:30:10,  1.82s/it]
 13%|█▎        | 1848/14725 [58:55<6:28:59,  1.81s/it]
 13%|█▎        | 1849/14725 
+0: {'loss': 0.8849, 'grad_norm': 0.8683864960522815, 'learning_rate': 2e-05, 'epoch': 0.13}
+0: {'loss': 0.8561, 'grad_norm': 0.7931060945079955, 'learning_rate': 2e-05, 'epoch': 0.13}
+0: [58:57<6:28:45,  1.81s/it]
 13%|█▎        | 1850/14725 [58:58<6:29:18,  1.81s/it]
                                                      

 13%|█▎        | 1850/14725 [58:58<6:29:18,  1.81s/it]
 13%|█▎        | 1851/14725 [59:00<6:36:12,  1.85s/it]
 13%|█▎        | 1852/14725 [59:02<6:33:04,  1.83s/it]
 13%|█▎        | 1853/14725 [59:04<6:31:04,  1.82s/it]
 13%|█▎        | 1854/14725 [59:06<6:29:36,  1.82s/it]
 13%|█▎        | 1855/14725 [59:08<6:32:57,  1.83s/it]
 13%|█▎        | 1856/14725 [59:09<6:30:45,  1.82s/it]
 13%|█▎        | 1857/14725 [59:11<6:29:39,  1.82s/it]
 13%|█▎        | 1858/14725 [59:13<6:29:47,  1.82s/it]
 13%|█▎        | 1859/14725 [59:15<6:28:13,  1.81s/it]
 13%|█▎        | 1860/14725 [59:17<6:26:10,  1.80s/it]
                                                      

 13%|█▎        | 1860/14725 [59:17<6:26:10,  1.80s/it]
 13%|█▎        | 1861/14725 [59:18<6:24:54,  1.80s/it]
 13%|█▎        | 1862/14725 [59:20<6:24:03,  1.79s/it]
+0: {'loss': 0.8931, 'grad_norm': 0.9231553920086579, 'learning_rate': 2e-05, 'epoch': 0.13}
+0:  13%|█▎        | 1863/14725 [59:22<6:27:16,  1.81s/it]
 13%|█▎        | 1864/14725 [59:24<6:28:03,  1.81s/it]
 13%|█▎        | 1865/14725 [59:26<6:25:47,  1.80s/it]
 13%|█▎        | 1866/14725 [59:27<6:27:51,  1.81s/it]
 13%|█▎        | 1867/14725 [59:29<6:35:36,  1.85s/it]
 13%|█▎        | 1868/14725 [59:31<6:35:47,  1.85s/it]
 13%|█▎        | 1869/14725 [59:33<6:33:12,  1.84s/it]
 13%|█▎        | 1870/14725 [59:35<6:32:35,  1.83s/it]
                                                      

 13%|█▎        | 1870/14725 [59:35<6:32:35,  1.83s/it]
 13%|█▎        | 1871/14725 [59:37<6:31:15,  1.83s/it]
 13%|█▎        | 1872/14725 [59:38<6:27:46,  1.81s/it]
 13%|█▎        | 1873/14725 [59:40<6:28:29,  1.81s/it]
 13%|█▎        | 1874/14725 [59:42<6:32:44,  1.83s/it]
 13%|█▎        | 1875/14725 [59:44<6:33:15,  1.84s/it]
 13%|█▎        | 1876/14725 [59:46<6:35:13,  1.85s/it]
 13%|█▎        | 1877/14725 [59:48<6:31:37,  1.83s/it]
 13%|█▎        | 187
+0: {'loss': 0.9096, 'grad_norm': 0.8805966765927816, 'learning_rate': 2e-05, 'epoch': 0.13}
+0: {'loss': 0.87, 'grad_norm': 0.8879304223937475, 'learning_rate': 2e-05, 'epoch': 0.13}
+0: 8/14725 [59:50<6:32:31,  1.83s/it]
 13%|█▎        | 1879/14725 [59:51<6:34:28,  1.84s/it]
 13%|█▎        | 1880/14725 [59:53<6:40:31,  1.87s/it]
                                                      

 13%|█▎        | 1880/14725 [59:53<6:40:31,  1.87s/it]
 13%|█▎        | 1881/14725 [59:55<6:36:36,  1.85s/it]
 13%|█▎        | 1882/14725 [59:57<6:35:07,  1.85s/it]
 13%|█▎        | 1883/14725 [59:59<6:36:09,  1.85s/it]
 13%|█▎        | 1884/14725 [1:00:01<6:35:29,  1.85s/it]
 13%|█▎        | 1885/14725 [1:00:02<6:32:41,  1.84s/it]
 13%|█▎        | 1886/14725 [1:00:04<6:35:23,  1.85s/it]
 13%|█▎        | 1887/14725 [1:00:06<6:29:57,  1.82s/it]
 13%|█▎        | 1888/14725 [1:00:08<6:26:53,  1.81s/it]
 13%|█▎        | 1889/14725 [1:00:10<6:24:43,  1.80s/it]
 13%|█▎        | 1890/14725 [1:00:11<6:26:53,  1.81s/it]
                                                        

 13%|█▎        | 1890/14725 [1:00:11<6:26:53,  1.81s/it]
 13%|█▎        | 1891/14725 [
+0: {'loss': 0.8689, 'grad_norm': 0.8198353583821275, 'learning_rate': 2e-05, 'epoch': 0.13}
+0: 1:00:13<6:28:00,  1.81s/it]
 13%|█▎        | 1892/14725 [1:00:15<6:27:34,  1.81s/it]
 13%|█▎        | 1893/14725 [1:00:17<6:26:26,  1.81s/it]
 13%|█▎        | 1894/14725 [1:00:19<6:24:05,  1.80s/it]
 13%|█▎        | 1895/14725 [1:00:20<6:23:45,  1.79s/it]
 13%|█▎        | 1896/14725 [1:00:22<6:23:02,  1.79s/it]
 13%|█▎        | 1897/14725 [1:00:24<6:22:07,  1.79s/it]
 13%|█▎        | 1898/14725 [1:00:26<6:28:08,  1.82s/it]
 13%|█▎        | 1899/14725 [1:00:28<6:30:49,  1.83s/it]
 13%|█▎        | 1900/14725 [1:00:30<6:29:47,  1.82s/it]
                                                        

 13%|█▎        | 1900/14725 [1:00:30<6:29:47,  1.82s/it]
 13%|█▎        | 1901/14725 [1:00:31<6:32:01,  1.83s/it]
 13%|█▎        | 1902/14725 [1:00:33<6:32:28,  1.84s/it]
 13%|█▎        | 1903/14725 [1:00:35<6:33:28,  1.84s/it]
 13%|█▎        | 1904/14725 [1:00:37<6:35:38,  1.85s/it]
 13%|█▎        | 1905/14725 [1:00:39<6:37:56,  1.86s/it]
 13%|█▎        | 19
+0: {'loss': 0.8907, 'grad_norm': 0.7908943971709059, 'learning_rate': 2e-05, 'epoch': 0.13}
+0: 06/14725 [1:00:41<6:39:53,  1.87s/it]
 13%|█▎        | 1907/14725 [1:00:43<6:33:46,  1.84s/it]
 13%|█▎        | 1908/14725 [1:00:44<6:28:48,  1.82s/it]
 13%|█▎        | 1909/14725 [1:00:46<6:30:32,  1.83s/it]
 13%|█▎        | 1910/14725 [1:00:48<6:28:11,  1.82s/it]
                                                        

 13%|█▎        | 1910/14725 [1:00:48<6:28:11,  1.82s/it]
 13%|█▎        | 1911/14725 [1:00:50<6:36:31,  1.86s/it]
 13%|█▎        | 1912/14725 [1:00:52<6:32:48,  1.84s/it]
 13%|█▎        | 1913/14725 [1:00:54<6:30:32,  1.83s/it]
 13%|█▎        | 1914/14725 [1:00:55<6:31:35,  1.83s/it]
 13%|█▎        | 1915/14725 [1:00:57<6:32:46,  1.84s/it]
 13%|█▎        | 1916/14725 [1:00:59<6:30:09,  1.83s/it]
 13%|█▎        | 1917/14725 [1:01:01<6:28:43,  1.82s/it]
 13%|█▎        | 1918/14725 [1:01:03<6:31:55,  1.84s/it]
 13%|█▎        | 1919/14725 [1:01:05<6:33:36,  1.84s/it]
 13%|█▎        | 1920/14725 [1:01:06<6:31:12,  1.83s/it]
             
+0: {'loss': 0.8929, 'grad_norm': 0.8295283793665107, 'learning_rate': 2e-05, 'epoch': 0.13}
+0: {'loss': 0.8753, 'grad_norm': 0.8914506775004732, 'learning_rate': 2e-05, 'epoch': 0.13}
+0:                                            

 13%|█▎        | 1920/14725 [1:01:06<6:31:12,  1.83s/it]
 13%|█▎        | 1921/14725 [1:01:08<6:31:27,  1.83s/it]
 13%|█▎        | 1922/14725 [1:01:10<6:34:13,  1.85s/it]
 13%|█▎        | 1923/14725 [1:01:12<6:31:22,  1.83s/it]
 13%|█▎        | 1924/14725 [1:01:14<6:34:45,  1.85s/it]
 13%|█▎        | 1925/14725 [1:01:16<6:32:08,  1.84s/it]
 13%|█▎        | 1926/14725 [1:01:17<6:29:41,  1.83s/it]
 13%|█▎        | 1927/14725 [1:01:19<6:28:58,  1.82s/it]
 13%|█▎        | 1928/14725 [1:01:21<6:34:23,  1.85s/it]
 13%|█▎        | 1929/14725 [1:01:23<6:37:11,  1.86s/it]
 13%|█▎        | 1930/14725 [1:01:25<6:33:42,  1.85s/it]
                                                        

 13%|█▎        | 1930/14725 [1:01:25<6:33:42,  1.85s/it]
 13%|█▎        | 1931/14725 [1:01:27<6:37:27,  1.86s/it]
 13%|█▎        | 1932/14725 [1:01:29<6:37:11,  1.86s/it]
 13%|█▎        | 1933/14725 [1:01:30<6:33:46,  1.85s/it]
 13%|�
+0: {'loss': 0.8966, 'grad_norm': 0.8651060093708194, 'learning_rate': 2e-05, 'epoch': 0.13}
+0: ��▎        | 1934/14725 [1:01:32<6:29:27,  1.83s/it]
 13%|█▎        | 1935/14725 [1:01:34<6:26:12,  1.81s/it]
 13%|█▎        | 1936/14725 [1:01:36<6:29:48,  1.83s/it]
 13%|█▎        | 1937/14725 [1:01:38<6:32:43,  1.84s/it]
 13%|█▎        | 1938/14725 [1:01:40<6:33:10,  1.84s/it]
 13%|█▎        | 1939/14725 [1:01:41<6:35:25,  1.86s/it]
 13%|█▎        | 1940/14725 [1:01:43<6:32:11,  1.84s/it]
                                                        

 13%|█▎        | 1940/14725 [1:01:43<6:32:11,  1.84s/it]
 13%|█▎        | 1941/14725 [1:01:45<6:27:42,  1.82s/it]
 13%|█▎        | 1942/14725 [1:01:47<6:28:01,  1.82s/it]
 13%|█▎        | 1943/14725 [1:01:49<6:41:22,  1.88s/it]
 13%|█▎        | 1944/14725 [1:01:51<6:40:23,  1.88s/it]
 13%|█▎        | 1945/14725 [1:01:53<6:34:56,  1.85s/it]
 13%|█▎        | 1946/14725 [1:01:54<6:32:55,  1.84s/it]
 13%|█▎        | 1947/14725 [1:01:56<6:31:58,  1.84s/it]
 13%|█▎        | 1948/14725 [1:01:58<6:27:56,  1.82s/
+0: {'loss': 0.9069, 'grad_norm': 0.9596799381921539, 'learning_rate': 2e-05, 'epoch': 0.13}
+0: {'loss': 0.8738, 'grad_norm': 0.8015498368431108, 'learning_rate': 2e-05, 'epoch': 0.13}
+0: it]
 13%|█▎        | 1949/14725 [1:02:00<6:24:22,  1.81s/it]
 13%|█▎        | 1950/14725 [1:02:02<6:24:06,  1.80s/it]
                                                        

 13%|█▎        | 1950/14725 [1:02:02<6:24:06,  1.80s/it]
 13%|█▎        | 1951/14725 [1:02:03<6:27:50,  1.82s/it]
 13%|█▎        | 1952/14725 [1:02:05<6:30:32,  1.83s/it]
 13%|█▎        | 1953/14725 [1:02:07<6:33:14,  1.85s/it]
 13%|█▎        | 1954/14725 [1:02:09<6:31:08,  1.84s/it]
 13%|█▎        | 1955/14725 [1:02:11<6:26:50,  1.82s/it]
 13%|█▎        | 1956/14725 [1:02:13<6:27:02,  1.82s/it]
 13%|█▎        | 1957/14725 [1:02:14<6:28:53,  1.83s/it]
 13%|█▎        | 1958/14725 [1:02:16<6:26:44,  1.82s/it]
 13%|█▎        | 1959/14725 [1:02:18<6:29:13,  1.83s/it]
 13%|█▎        | 1960/14725 [1:02:20<6:29:13,  1.83s/it]
                                                        

 13%|█▎        | 1960/14725 [1:02:20<6:29:13,  1.83s/it]
 13%|█▎        | 1961/14725 [1:02:22<6:25:55, 
+0: {'loss': 0.8885, 'grad_norm': 0.8813575955861994, 'learning_rate': 2e-05, 'epoch': 0.13}
+0:  1.81s/it]
 13%|█▎        | 1962/14725 [1:02:23<6:23:41,  1.80s/it]
 13%|█▎        | 1963/14725 [1:02:25<6:26:25,  1.82s/it]
 13%|█▎        | 1964/14725 [1:02:27<6:30:25,  1.84s/it]
 13%|█▎        | 1965/14725 [1:02:29<6:30:55,  1.84s/it]
 13%|█▎        | 1966/14725 [1:02:31<6:26:28,  1.82s/it]
 13%|█▎        | 1967/14725 [1:02:33<6:28:34,  1.83s/it]
 13%|█▎        | 1968/14725 [1:02:34<6:26:58,  1.82s/it]
 13%|█▎        | 1969/14725 [1:02:36<6:25:55,  1.82s/it]
 13%|█▎        | 1970/14725 [1:02:38<6:24:24,  1.81s/it]
                                                        

 13%|█▎        | 1970/14725 [1:02:38<6:24:24,  1.81s/it]
 13%|█▎        | 1971/14725 [1:02:40<6:23:37,  1.80s/it]
 13%|█▎        | 1972/14725 [1:02:42<6:28:43,  1.83s/it]
 13%|█▎        | 1973/14725 [1:02:43<6:25:18,  1.81s/it]
 13%|█▎        | 1974/14725 [1:02:45<6:24:11,  1.81s/it]
 13%|█▎        | 1975/14725 [1:02:47<6:22:57,  1.80s/it]
 13%|█▎        | 1976/14725 [1:02:49
+0: {'loss': 0.8796, 'grad_norm': 0.834385476197543, 'learning_rate': 2e-05, 'epoch': 0.13}
+0: <6:24:37,  1.81s/it]
 13%|█▎        | 1977/14725 [1:02:51<6:22:15,  1.80s/it]
 13%|█▎        | 1978/14725 [1:02:52<6:19:58,  1.79s/it]
 13%|█▎        | 1979/14725 [1:02:54<6:19:13,  1.79s/it]
 13%|█▎        | 1980/14725 [1:02:56<6:19:11,  1.79s/it]
                                                        

 13%|█▎        | 1980/14725 [1:02:56<6:19:11,  1.79s/it]
 13%|█▎        | 1981/14725 [1:02:58<6:25:40,  1.82s/it]
 13%|█▎        | 1982/14725 [1:03:00<6:28:13,  1.83s/it]
 13%|█▎        | 1983/14725 [1:03:02<6:27:39,  1.83s/it]
 13%|█▎        | 1984/14725 [1:03:03<6:25:35,  1.82s/it]
 13%|█▎        | 1985/14725 [1:03:05<6:24:14,  1.81s/it]
 13%|█▎        | 1986/14725 [1:03:07<6:21:37,  1.80s/it]
 13%|█▎        | 1987/14725 [1:03:09<6:21:51,  1.80s/it]
 14%|█▎        | 1988/14725 [1:03:11<6:22:04,  1.80s/it]
 14%|█▎        | 1989/14725 [1:03:12<6:21:12,  1.80s/it]
 14%|█▎        | 1990/14725 [1:03:14<6:21:58,  1.80s/it]
                              
+0: {'loss': 0.8653, 'grad_norm': 0.9305717091872017, 'learning_rate': 2e-05, 'epoch': 0.14}
+0: {'loss': 0.8965, 'grad_norm': 0.788477293894533, 'learning_rate': 2e-05, 'epoch': 0.14}
+0:                           

 14%|█▎        | 1990/14725 [1:03:14<6:21:58,  1.80s/it]
 14%|█▎        | 1991/14725 [1:03:16<6:22:00,  1.80s/it]
 14%|█▎        | 1992/14725 [1:03:18<6:20:19,  1.79s/it]
 14%|█▎        | 1993/14725 [1:03:19<6:19:35,  1.79s/it]
 14%|█▎        | 1994/14725 [1:03:21<6:22:28,  1.80s/it]
 14%|█▎        | 1995/14725 [1:03:23<6:23:00,  1.81s/it]
 14%|█▎        | 1996/14725 [1:03:25<6:31:11,  1.84s/it]
 14%|█▎        | 1997/14725 [1:03:27<6:28:47,  1.83s/it]
 14%|█▎        | 1998/14725 [1:03:29<6:27:18,  1.83s/it]
 14%|█▎        | 1999/14725 [1:03:30<6:25:04,  1.82s/it]
 14%|█▎        | 2000/14725 [1:03:32<6:35:46,  1.87s/it]
                                                        

 14%|█▎        | 2000/14725 [1:03:32<6:35:46,  1.87s/it]
 14%|█▎        | 2001/14725 [1:03:34<6:35:01,  1.86s/it]
 14%|█▎        | 2002/14725 [1:03:36<6:31:52,  1.85s/it]
 14%|█▎        | 2003/14725 [1:03:38<6:29:03,  1.83s/it]
 14%|█▎        | 20
+0: {'loss': 0.885, 'grad_norm': 0.8363277925082019, 'learning_rate': 2e-05, 'epoch': 0.14}
+0: 04/14725 [1:03:40<6:26:17,  1.82s/it]
 14%|█▎        | 2005/14725 [1:03:42<6:26:46,  1.82s/it]
 14%|█▎        | 2006/14725 [1:03:43<6:24:03,  1.81s/it]
 14%|█▎        | 2007/14725 [1:03:45<6:26:29,  1.82s/it]
 14%|█▎        | 2008/14725 [1:03:47<6:26:34,  1.82s/it]
 14%|█▎        | 2009/14725 [1:03:49<6:26:01,  1.82s/it]
 14%|█▎        | 2010/14725 [1:03:51<6:23:57,  1.81s/it]
                                                        

 14%|█▎        | 2010/14725 [1:03:51<6:23:57,  1.81s/it]
 14%|█▎        | 2011/14725 [1:03:52<6:23:08,  1.81s/it]
 14%|█▎        | 2012/14725 [1:03:54<6:25:07,  1.82s/it]
 14%|█▎        | 2013/14725 [1:03:56<6:24:28,  1.81s/it]
 14%|█▎        | 2014/14725 [1:03:58<6:23:39,  1.81s/it]
 14%|█▎        | 2015/14725 [1:04:00<6:22:45,  1.81s/it]
 14%|█▎        | 2016/14725 [1:04:02<6:30:55,  1.85s/it]
 14%|█▎        | 2017/14725 [1:04:03<6:27:56,  1.83s/it]
 14%|█▎        | 2018/14725 [1:04:05<6:23:53,  1.81s/it]
 14%|█▎  
+0: {'loss': 0.8846, 'grad_norm': 0.8346312151440024, 'learning_rate': 2e-05, 'epoch': 0.14}
+0: {'loss': 0.8799, 'grad_norm': 0.817869027490219, 'learning_rate': 2e-05, 'epoch': 0.14}
+0:       | 2019/14725 [1:04:07<6:25:28,  1.82s/it]
 14%|█▎        | 2020/14725 [1:04:09<6:23:14,  1.81s/it]
                                                        

 14%|█▎        | 2020/14725 [1:04:09<6:23:14,  1.81s/it]
 14%|█▎        | 2021/14725 [1:04:11<6:28:02,  1.83s/it]
 14%|█▎        | 2022/14725 [1:04:12<6:24:49,  1.82s/it]
 14%|█▎        | 2023/14725 [1:04:14<6:23:15,  1.81s/it]
 14%|█▎        | 2024/14725 [1:04:16<6:24:18,  1.82s/it]
 14%|█▍        | 2025/14725 [1:04:18<6:21:13,  1.80s/it]
 14%|█▍        | 2026/14725 [1:04:20<6:20:27,  1.80s/it]
 14%|█▍        | 2027/14725 [1:04:22<6:27:54,  1.83s/it]
 14%|█▍        | 2028/14725 [1:04:23<6:27:44,  1.83s/it]
 14%|█▍        | 2029/14725 [1:04:25<6:26:31,  1.83s/it]
 14%|█▍        | 2030/14725 [1:04:27<6:25:16,  1.82s/it]
                                                        

 14%|█▍        | 2030/14725 [1:04:27<6:25:16,  1.82s/it]
 14%|█▍        | 2031/14725 [1:04:29<6:28:27,  1.84s/it]
 14%|�
+0: {'loss': 0.9046, 'grad_norm': 0.8018256280322785, 'learning_rate': 2e-05, 'epoch': 0.14}
+0: ��▍        | 2032/14725 [1:04:31<6:30:13,  1.84s/it]
 14%|█▍        | 2033/14725 [1:04:33<6:29:21,  1.84s/it]
 14%|█▍        | 2034/14725 [1:04:34<6:30:30,  1.85s/it]
 14%|█▍        | 2035/14725 [1:04:36<6:29:19,  1.84s/it]
 14%|█▍        | 2036/14725 [1:04:38<6:30:26,  1.85s/it]
 14%|█▍        | 2037/14725 [1:04:40<6:34:19,  1.86s/it]
 14%|█▍        | 2038/14725 [1:04:42<6:30:15,  1.85s/it]
 14%|█▍        | 2039/14725 [1:04:44<6:26:33,  1.83s/it]
 14%|█▍        | 2040/14725 [1:04:45<6:27:26,  1.83s/it]
                                                        

 14%|█▍        | 2040/14725 [1:04:45<6:27:26,  1.83s/it]
 14%|█▍        | 2041/14725 [1:04:47<6:28:39,  1.84s/it]
 14%|█▍        | 2042/14725 [1:04:49<6:25:08,  1.82s/it]
 14%|█▍        | 2043/14725 [1:04:51<6:24:12,  1.82s/it]
 14%|█▍        | 2044/14725 [1:04:53<6:22:06,  1.81s/it]
 14%|█▍        | 2045/14725 [1:04:54<6:21:56,  1.81s/it]
 14%|█▍        | 2046/14725 [1:04:56<6:19:35,  1.80s/
+0: {'loss': 0.8747, 'grad_norm': 0.807023208403021, 'learning_rate': 2e-05, 'epoch': 0.14}
+0: it]
 14%|█▍        | 2047/14725 [1:04:58<6:20:58,  1.80s/it]
 14%|█▍        | 2048/14725 [1:05:00<6:20:31,  1.80s/it]
 14%|█▍        | 2049/14725 [1:05:02<6:22:05,  1.81s/it]
 14%|█▍        | 2050/14725 [1:05:04<6:33:24,  1.86s/it]
                                                        

 14%|█▍        | 2050/14725 [1:05:04<6:33:24,  1.86s/it]
 14%|█▍        | 2051/14725 [1:05:06<6:34:12,  1.87s/it]
 14%|█▍        | 2052/14725 [1:05:07<6:28:27,  1.84s/it]
 14%|█▍        | 2053/14725 [1:05:09<6:27:31,  1.83s/it]
 14%|█▍        | 2054/14725 [1:05:11<6:25:54,  1.83s/it]
 14%|█▍        | 2055/14725 [1:05:13<6:23:32,  1.82s/it]
 14%|█▍        | 2056/14725 [1:05:15<6:25:21,  1.83s/it]
 14%|█▍        | 2057/14725 [1:05:16<6:26:41,  1.83s/it]
 14%|█▍        | 2058/14725 [1:05:18<6:23:38,  1.82s/it]
 14%|█▍        | 2059/14725 [1:05:20<6:37:13,  1.88s/it]
 14%|█▍        | 2060/14725 [1:05:22<6:34:07,  1.87s/it]
                                               
+0: {'loss': 0.8828, 'grad_norm': 0.8066333158587135, 'learning_rate': 2e-05, 'epoch': 0.14}
+0: {'loss': 0.8691, 'grad_norm': 0.7892768595029754, 'learning_rate': 2e-05, 'epoch': 0.14}
+0:          

 14%|█▍        | 2060/14725 [1:05:22<6:34:07,  1.87s/it]
 14%|█▍        | 2061/14725 [1:05:24<6:30:08,  1.85s/it]
 14%|█▍        | 2062/14725 [1:05:26<6:44:12,  1.92s/it]
 14%|█▍        | 2063/14725 [1:05:28<6:36:58,  1.88s/it]
 14%|█▍        | 2064/14725 [1:05:30<6:34:05,  1.87s/it]
 14%|█▍        | 2065/14725 [1:05:31<6:29:20,  1.85s/it]
 14%|█▍        | 2066/14725 [1:05:33<6:35:57,  1.88s/it]
 14%|█▍        | 2067/14725 [1:05:35<6:31:44,  1.86s/it]
 14%|█▍        | 2068/14725 [1:05:37<6:33:10,  1.86s/it]
 14%|█▍        | 2069/14725 [1:05:39<6:44:32,  1.92s/it]
 14%|█▍        | 2070/14725 [1:05:41<6:50:55,  1.95s/it]
                                                        

 14%|█▍        | 2070/14725 [1:05:41<6:50:55,  1.95s/it]
 14%|█▍        | 2071/14725 [1:05:43<6:52:12,  1.95s/it]
 14%|█▍        | 2072/14725 [1:05:45<7:00:23,  1.99s/it]
 14%|█▍        | 2073/14725 [1:05:47<6:52:47,  1.96s/it]
 14%|█▍        | 2074/14725 [1:05:49
+0: {'loss': 0.9081, 'grad_norm': 0.8634050983050812, 'learning_rate': 2e-05, 'epoch': 0.14}
+0: <6:43:53,  1.92s/it]
 14%|█▍        | 2075/14725 [1:05:51<6:56:15,  1.97s/it]
 14%|█▍        | 2076/14725 [1:05:53<6:46:17,  1.93s/it]
 14%|█▍        | 2077/14725 [1:05:55<6:44:52,  1.92s/it]
 14%|█▍        | 2078/14725 [1:05:57<7:02:15,  2.00s/it]
 14%|█▍        | 2079/14725 [1:06:00<7:47:02,  2.22s/it]
 14%|█▍        | 2080/14725 [1:06:03<8:56:10,  2.54s/it]
                                                        

 14%|█▍        | 2080/14725 [1:06:03<8:56:10,  2.54s/it]
 14%|█▍        | 2081/14725 [1:06:05<8:12:36,  2.34s/it]
 14%|█▍        | 2082/14725 [1:06:07<7:43:42,  2.20s/it]
 14%|█▍        | 2083/14725 [1:06:09<7:34:30,  2.16s/it]
 14%|█▍        | 2084/14725 [1:06:10<7:10:42,  2.04s/it]
 14%|█▍        | 2085/14725 [1:06:12<6:53:34,  1.96s/it]
 14%|█▍        | 2086/14725 [1:06:14<6:46:42,  1.93s/it]
 14%|█▍        | 2087/14725 [1:06:16<6:39:56,  1.90s/it]
 14%|█▍        | 2088/14725 [1:06:18<6:31:56,  1.86s/it]
 14%|█▍        | 2089/1472
+0: {'loss': 0.8874, 'grad_norm': 0.9955053215249907, 'learning_rate': 2e-05, 'epoch': 0.14}
+0: {'loss': 0.8905, 'grad_norm': 0.9646887646962127, 'learning_rate': 2e-05, 'epoch': 0.14}
+0: 5 [1:06:20<6:30:03,  1.85s/it]
 14%|█▍        | 2090/14725 [1:06:21<6:23:32,  1.82s/it]
                                                        

 14%|█▍        | 2090/14725 [1:06:21<6:23:32,  1.82s/it]
 14%|█▍        | 2091/14725 [1:06:23<6:35:01,  1.88s/it]
 14%|█▍        | 2092/14725 [1:06:25<6:32:25,  1.86s/it]
 14%|█▍        | 2093/14725 [1:06:27<7:00:55,  2.00s/it]
 14%|█▍        | 2094/14725 [1:06:29<6:47:43,  1.94s/it]
 14%|█▍        | 2095/14725 [1:06:31<6:38:31,  1.89s/it]
 14%|█▍        | 2096/14725 [1:06:33<6:36:30,  1.88s/it]
 14%|█▍        | 2097/14725 [1:06:35<6:29:36,  1.85s/it]
 14%|█▍        | 2098/14725 [1:06:36<6:24:16,  1.83s/it]
 14%|█▍        | 2099/14725 [1:06:38<6:24:13,  1.83s/it]
 14%|█▍        | 2100/14725 [1:06:40<6:19:40,  1.80s/it]
                                                        

 14%|█▍        | 2100/14725 [1:06:40<6:19:40,  1.80s/it]
 14%|█▍        | 2101/14725 [1:06:42<6:22:20,  1.82s/it]
 14%|█▍        | 21
+0: {'loss': 0.8582, 'grad_norm': 0.9200675323552091, 'learning_rate': 2e-05, 'epoch': 0.14}
+0: 02/14725 [1:06:44<6:22:40,  1.82s/it]
 14%|█▍        | 2103/14725 [1:06:45<6:20:39,  1.81s/it]
 14%|█▍        | 2104/14725 [1:06:47<6:26:38,  1.84s/it]
 14%|█▍        | 2105/14725 [1:06:49<6:25:51,  1.83s/it]
 14%|█▍        | 2106/14725 [1:06:51<6:29:21,  1.85s/it]
 14%|█▍        | 2107/14725 [1:06:53<6:25:29,  1.83s/it]
 14%|█▍        | 2108/14725 [1:06:55<6:27:32,  1.84s/it]
 14%|█▍        | 2109/14725 [1:06:57<6:31:08,  1.86s/it]
 14%|█▍        | 2110/14725 [1:06:58<6:25:39,  1.83s/it]
                                                        

 14%|█▍        | 2110/14725 [1:06:58<6:25:39,  1.83s/it]
 14%|█▍        | 2111/14725 [1:07:00<6:24:06,  1.83s/it]
 14%|█▍        | 2112/14725 [1:07:02<6:25:10,  1.83s/it]
 14%|█▍        | 2113/14725 [1:07:04<6:21:37,  1.82s/it]
 14%|█▍        | 2114/14725 [1:07:06<6:19:20,  1.80s/it]
 14%|█▍        | 2115/14725 [1:07:07<6:19:36,  1.81s/it]
 14%|█▍        | 2116/14725 [1:07:09<6:19:39,  1.81s/it]
 14%|█▍  
+0: {'loss': 0.8773, 'grad_norm': 0.9050627900650766, 'learning_rate': 2e-05, 'epoch': 0.14}
+0: {'loss': 0.8628, 'grad_norm': 0.816790586958817, 'learning_rate': 2e-05, 'epoch': 0.14}
+0:       | 2117/14725 [1:07:11<6:22:21,  1.82s/it]
 14%|█▍        | 2118/14725 [1:07:13<6:18:19,  1.80s/it]
 14%|█▍        | 2119/14725 [1:07:15<6:17:31,  1.80s/it]
 14%|█▍        | 2120/14725 [1:07:16<6:21:58,  1.82s/it]
                                                        

 14%|█▍        | 2120/14725 [1:07:16<6:21:58,  1.82s/it]
 14%|█▍        | 2121/14725 [1:07:18<6:21:10,  1.81s/it]
 14%|█▍        | 2122/14725 [1:07:20<6:17:52,  1.80s/it]
 14%|█▍        | 2123/14725 [1:07:22<6:17:53,  1.80s/it]
 14%|█▍        | 2124/14725 [1:07:24<6:20:54,  1.81s/it]
 14%|█▍        | 2125/14725 [1:07:25<6:17:24,  1.80s/it]
 14%|█▍        | 2126/14725 [1:07:27<6:15:14,  1.79s/it]
 14%|█▍        | 2127/14725 [1:07:29<6:18:10,  1.80s/it]
 14%|█▍        | 2128/14725 [1:07:31<6:19:48,  1.81s/it]
 14%|█▍        | 2129/14725 [1:07:33<6:28:38,  1.85s/it]
 14%|█▍        | 2130/14725 [1:07:35<6:27:16,  1.84s/it]
                                                        

 14%|�
+0: {'loss': 0.8785, 'grad_norm': 0.8083868094537943, 'learning_rate': 2e-05, 'epoch': 0.15}
+0: ��▍        | 2130/14725 [1:07:35<6:27:16,  1.84s/it]
 14%|█▍        | 2131/14725 [1:07:36<6:22:57,  1.82s/it]
 14%|█▍        | 2132/14725 [1:07:38<6:20:59,  1.82s/it]
 14%|█▍        | 2133/14725 [1:07:40<6:19:25,  1.81s/it]
 14%|█▍        | 2134/14725 [1:07:42<6:17:41,  1.80s/it]
 14%|█▍        | 2135/14725 [1:07:44<6:20:04,  1.81s/it]
 15%|█▍        | 2136/14725 [1:07:45<6:18:48,  1.81s/it]
 15%|█▍        | 2137/14725 [1:07:47<6:19:44,  1.81s/it]
 15%|█▍        | 2138/14725 [1:07:49<6:19:06,  1.81s/it]
 15%|█▍        | 2139/14725 [1:07:51<6:18:07,  1.80s/it]
 15%|█▍        | 2140/14725 [1:07:53<6:20:04,  1.81s/it]
                                                        

 15%|█▍        | 2140/14725 [1:07:53<6:20:04,  1.81s/it]
 15%|█▍        | 2141/14725 [1:07:54<6:17:49,  1.80s/it]
 15%|█▍        | 2142/14725 [1:07:56<6:17:49,  1.80s/it]
 15%|█▍        | 2143/14725 [1:07:58<6:17:22,  1.80s/it]
 15%|█▍        | 2144/14725 [1:08:00<6:14:57,  1.79s/
+0: {'loss': 0.8779, 'grad_norm': 0.8955066440295506, 'learning_rate': 2e-05, 'epoch': 0.15}
+0: it]
 15%|█▍        | 2145/14725 [1:08:02<6:14:58,  1.79s/it]
 15%|█▍        | 2146/14725 [1:08:03<6:20:35,  1.82s/it]
 15%|█▍        | 2147/14725 [1:08:05<6:17:25,  1.80s/it]
 15%|█▍        | 2148/14725 [1:08:07<6:16:29,  1.80s/it]
 15%|█▍        | 2149/14725 [1:08:09<6:19:34,  1.81s/it]
 15%|█▍        | 2150/14725 [1:08:11<6:15:39,  1.79s/it]
                                                        

 15%|█▍        | 2150/14725 [1:08:11<6:15:39,  1.79s/it]
 15%|█▍        | 2151/14725 [1:08:12<6:16:54,  1.80s/it]
 15%|█▍        | 2152/14725 [1:08:14<6:21:33,  1.82s/it]
 15%|█▍        | 2153/14725 [1:08:16<6:19:07,  1.81s/it]
 15%|█▍        | 2154/14725 [1:08:18<6:15:43,  1.79s/it]
 15%|█▍        | 2155/14725 [1:08:20<6:18:35,  1.81s/it]
 15%|█▍        | 2156/14725 [1:08:21<6:16:40,  1.80s/it]
 15%|█▍        | 2157/14725 [1:08:23<6:18:34,  1.81s/it]
 15%|█▍        | 2158/14725 [1:08:25<6:21:10,  1.82s/it]
 15%|█▍        | 2159/14725 [1:08:27<6:22:1
+0: {'loss': 0.8784, 'grad_norm': 0.805967681802712, 'learning_rate': 2e-05, 'epoch': 0.15}
+0: {'loss': 0.8873, 'grad_norm': 0.9458717959567946, 'learning_rate': 2e-05, 'epoch': 0.15}
+0: 0,  1.82s/it]
 15%|█▍        | 2160/14725 [1:08:29<6:18:29,  1.81s/it]
                                                        

 15%|█▍        | 2160/14725 [1:08:29<6:18:29,  1.81s/it]
 15%|█▍        | 2161/14725 [1:08:31<6:17:52,  1.80s/it]
 15%|█▍        | 2162/14725 [1:08:32<6:21:28,  1.82s/it]
 15%|█▍        | 2163/14725 [1:08:34<6:21:41,  1.82s/it]
 15%|█▍        | 2164/14725 [1:08:36<6:18:46,  1.81s/it]
 15%|█▍        | 2165/14725 [1:08:38<6:26:19,  1.85s/it]
 15%|█▍        | 2166/14725 [1:08:40<6:23:40,  1.83s/it]
 15%|█▍        | 2167/14725 [1:08:42<6:23:11,  1.83s/it]
 15%|█▍        | 2168/14725 [1:08:44<6:32:46,  1.88s/it]
 15%|█▍        | 2169/14725 [1:08:45<6:29:25,  1.86s/it]
 15%|█▍        | 2170/14725 [1:08:47<6:27:37,  1.85s/it]
                                                        

 15%|█▍        | 2170/14725 [1:08:47<6:27:37,  1.85s/it]
 15%|█▍        | 2171/14725 [1:08:49<6:23:08,  1.83s/it]
 15%|█▍        | 2172/14725 [1:08:51
+0: {'loss': 0.8785, 'grad_norm': 0.8353912501706671, 'learning_rate': 2e-05, 'epoch': 0.15}
+0: <6:20:36,  1.82s/it]
 15%|█▍        | 2173/14725 [1:08:53<6:25:52,  1.84s/it]
 15%|█▍        | 2174/14725 [1:08:54<6:20:50,  1.82s/it]
 15%|█▍        | 2175/14725 [1:08:56<6:24:27,  1.84s/it]
 15%|█▍        | 2176/14725 [1:08:58<6:27:45,  1.85s/it]
 15%|█▍        | 2177/14725 [1:09:00<6:26:48,  1.85s/it]
 15%|█▍        | 2178/14725 [1:09:02<6:23:22,  1.83s/it]
 15%|█▍        | 2179/14725 [1:09:04<6:24:29,  1.84s/it]
 15%|█▍        | 2180/14725 [1:09:06<6:24:19,  1.84s/it]
                                                        

 15%|█▍        | 2180/14725 [1:09:06<6:24:19,  1.84s/it]
 15%|█▍        | 2181/14725 [1:09:07<6:20:56,  1.82s/it]
 15%|█▍        | 2182/14725 [1:09:09<6:20:04,  1.82s/it]
 15%|█▍        | 2183/14725 [1:09:11<6:21:56,  1.83s/it]
 15%|█▍        | 2184/14725 [1:09:13<6:21:19,  1.82s/it]
 15%|█▍        | 2185/14725 [1:09:15<6:19:00,  1.81s/it]
 15%|█▍        | 2186/14725 [1:09:16<6:20:06,  1.82s/it]
 15%|█▍        | 2187/1472
+0: {'loss': 0.8916, 'grad_norm': 0.7786184502420562, 'learning_rate': 2e-05, 'epoch': 0.15}
+0: {'loss': 0.8758, 'grad_norm': 0.9212985770530212, 'learning_rate': 2e-05, 'epoch': 0.15}
+0: 5 [1:09:18<6:22:13,  1.83s/it]
 15%|█▍        | 2188/14725 [1:09:20<6:17:24,  1.81s/it]
 15%|█▍        | 2189/14725 [1:09:22<6:20:30,  1.82s/it]
 15%|█▍        | 2190/14725 [1:09:24<6:16:55,  1.80s/it]
                                                        

 15%|█▍        | 2190/14725 [1:09:24<6:16:55,  1.80s/it]
 15%|█▍        | 2191/14725 [1:09:25<6:14:50,  1.79s/it]
 15%|█▍        | 2192/14725 [1:09:27<6:19:03,  1.81s/it]
 15%|█▍        | 2193/14725 [1:09:29<6:26:22,  1.85s/it]
 15%|█▍        | 2194/14725 [1:09:31<6:23:21,  1.84s/it]
 15%|█▍        | 2195/14725 [1:09:33<6:20:16,  1.82s/it]
 15%|█▍        | 2196/14725 [1:09:35<6:20:32,  1.82s/it]
 15%|█▍        | 2197/14725 [1:09:36<6:17:39,  1.81s/it]
 15%|█▍        | 2198/14725 [1:09:38<6:18:02,  1.81s/it]
 15%|█▍        | 2199/14725 [1:09:40<6:18:07,  1.81s/it]
 15%|█▍        | 2200/14725 [1:09:42<6:20:51,  1.82s/it]
                                                        

 15%|█▍        | 22
+0: {'loss': 0.8543, 'grad_norm': 0.8250378345106958, 'learning_rate': 2e-05, 'epoch': 0.15}
+0: 00/14725 [1:09:42<6:20:51,  1.82s/it]
 15%|█▍        | 2201/14725 [1:09:44<6:19:33,  1.82s/it]
 15%|█▍        | 2202/14725 [1:09:46<6:19:45,  1.82s/it]
 15%|█▍        | 2203/14725 [1:09:47<6:15:23,  1.80s/it]
 15%|█▍        | 2204/14725 [1:09:49<6:18:52,  1.82s/it]
 15%|█▍        | 2205/14725 [1:09:51<6:18:07,  1.81s/it]
 15%|█▍        | 2206/14725 [1:09:53<6:15:18,  1.80s/it]
 15%|█▍        | 2207/14725 [1:09:55<6:17:57,  1.81s/it]
 15%|█▍        | 2208/14725 [1:09:56<6:14:36,  1.80s/it]
 15%|█▌        | 2209/14725 [1:09:58<6:16:34,  1.81s/it]
 15%|█▌        | 2210/14725 [1:10:00<6:14:14,  1.79s/it]
                                                        

 15%|█▌        | 2210/14725 [1:10:00<6:14:14,  1.79s/it]
 15%|█▌        | 2211/14725 [1:10:02<6:17:50,  1.81s/it]
 15%|█▌        | 2212/14725 [1:10:04<6:14:32,  1.80s/it]
 15%|█▌        | 2213/14725 [1:10:05<6:16:32,  1.81s/it]
 15%|█▌        | 2214/14725 [1:10:07<6:16:12,  1.80s/it]
 15%|█▌  
+0: {'loss': 0.8863, 'grad_norm': 0.8285311033027637, 'learning_rate': 2e-05, 'epoch': 0.15}
+0:       | 2215/14725 [1:10:09<6:18:53,  1.82s/it]
 15%|█▌        | 2216/14725 [1:10:11<6:15:00,  1.80s/it]
 15%|█▌        | 2217/14725 [1:10:13<6:12:19,  1.79s/it]
 15%|█▌        | 2218/14725 [1:10:14<6:14:52,  1.80s/it]
 15%|█▌        | 2219/14725 [1:10:16<6:17:29,  1.81s/it]
 15%|█▌        | 2220/14725 [1:10:18<6:15:19,  1.80s/it]
                                                        

 15%|█▌        | 2220/14725 [1:10:18<6:15:19,  1.80s/it]
 15%|█▌        | 2221/14725 [1:10:20<6:18:58,  1.82s/it]
 15%|█▌        | 2222/14725 [1:10:22<6:14:34,  1.80s/it]
 15%|█▌        | 2223/14725 [1:10:23<6:19:03,  1.82s/it]
 15%|█▌        | 2224/14725 [1:10:25<6:15:58,  1.80s/it]
 15%|█▌        | 2225/14725 [1:10:27<6:16:14,  1.81s/it]
 15%|█▌        | 2226/14725 [1:10:29<6:17:40,  1.81s/it]
 15%|█▌        | 2227/14725 [1:10:31<6:15:25,  1.80s/it]
 15%|█▌        | 2228/14725 [1:10:32<6:13:31,  1.79s/it]
 15%|█▌        | 2229/14725 [1:10:34<6:23:31,  1.84s/it]
 15
+0: {'loss': 0.8748, 'grad_norm': 0.8691302078187372, 'learning_rate': 2e-05, 'epoch': 0.15}
+0: {'loss': 0.8948, 'grad_norm': 0.8229830831931124, 'learning_rate': 2e-05, 'epoch': 0.15}
+0: %|█▌        | 2230/14725 [1:10:36<6:26:44,  1.86s/it]
                                                        

 15%|█▌        | 2230/14725 [1:10:36<6:26:44,  1.86s/it]
 15%|█▌        | 2231/14725 [1:10:38<6:29:55,  1.87s/it]
 15%|█▌        | 2232/14725 [1:10:40<6:31:40,  1.88s/it]
 15%|█▌        | 2233/14725 [1:10:42<6:32:27,  1.89s/it]
 15%|█▌        | 2234/14725 [1:10:44<6:31:37,  1.88s/it]
 15%|█▌        | 2235/14725 [1:10:46<6:30:16,  1.87s/it]
 15%|█▌        | 2236/14725 [1:10:47<6:23:44,  1.84s/it]
 15%|█▌        | 2237/14725 [1:10:49<6:20:44,  1.83s/it]
 15%|█▌        | 2238/14725 [1:10:51<6:22:04,  1.84s/it]
 15%|█▌        | 2239/14725 [1:10:53<6:19:40,  1.82s/it]
 15%|█▌        | 2240/14725 [1:10:55<6:17:14,  1.81s/it]
                                                        

 15%|█▌        | 2240/14725 [1:10:55<6:17:14,  1.81s/it]
 15%|█▌        | 2241/14725 [1:10:56<6:16:37,  1.81s/it]
 15%|█▌        | 2242/14725 [1:10:58<6:17:35,  1.81s/
+0: {'loss': 0.8924, 'grad_norm': 0.7816244969892461, 'learning_rate': 2e-05, 'epoch': 0.15}
+0: it]
 15%|█▌        | 2243/14725 [1:11:00<6:17:50,  1.82s/it]
 15%|█▌        | 2244/14725 [1:11:02<6:18:57,  1.82s/it]
 15%|█▌        | 2245/14725 [1:11:04<6:14:51,  1.80s/it]
 15%|█▌        | 2246/14725 [1:11:06<6:17:08,  1.81s/it]
 15%|█▌        | 2247/14725 [1:11:07<6:16:32,  1.81s/it]
 15%|█▌        | 2248/14725 [1:11:09<6:14:25,  1.80s/it]
 15%|█▌        | 2249/14725 [1:11:11<6:11:41,  1.79s/it]
 15%|█▌        | 2250/14725 [1:11:13<6:15:07,  1.80s/it]
                                                        

 15%|█▌        | 2250/14725 [1:11:13<6:15:07,  1.80s/it]
 15%|█▌        | 2251/14725 [1:11:15<6:13:51,  1.80s/it]
 15%|█▌        | 2252/14725 [1:11:16<6:14:58,  1.80s/it]
 15%|█▌        | 2253/14725 [1:11:18<6:12:53,  1.79s/it]
 15%|█▌        | 2254/14725 [1:11:20<6:14:29,  1.80s/it]
 15%|█▌        | 2255/14725 [1:11:22<6:11:14,  1.79s/it]
 15%|█▌        | 2256/14725 [1:11:24<6:15:48,  1.81s/it]
 15%|█▌        | 2257/14725 [1:11:25<6:16:5
+0: {'loss': 0.8867, 'grad_norm': 0.9273543823884622, 'learning_rate': 2e-05, 'epoch': 0.15}
+0: {'loss': 0.8759, 'grad_norm': 0.7540076102636767, 'learning_rate': 2e-05, 'epoch': 0.15}
+0: 8,  1.81s/it]
 15%|█▌        | 2258/14725 [1:11:27<6:17:29,  1.82s/it]
 15%|█▌        | 2259/14725 [1:11:29<6:16:47,  1.81s/it]
 15%|█▌        | 2260/14725 [1:11:31<6:18:15,  1.82s/it]
                                                        

 15%|█▌        | 2260/14725 [1:11:31<6:18:15,  1.82s/it]
 15%|█▌        | 2261/14725 [1:11:33<6:20:57,  1.83s/it]
 15%|█▌        | 2262/14725 [1:11:35<6:19:22,  1.83s/it]
 15%|█▌        | 2263/14725 [1:11:36<6:18:43,  1.82s/it]
 15%|█▌        | 2264/14725 [1:11:38<6:14:46,  1.80s/it]
 15%|█▌        | 2265/14725 [1:11:40<6:19:02,  1.83s/it]
 15%|█▌        | 2266/14725 [1:11:42<6:21:22,  1.84s/it]
 15%|█▌        | 2267/14725 [1:11:44<6:16:30,  1.81s/it]
 15%|█▌        | 2268/14725 [1:11:45<6:17:43,  1.82s/it]
 15%|█▌        | 2269/14725 [1:11:47<6:20:18,  1.83s/it]
 15%|█▌        | 2270/14725 [1:11:49<6:16:53,  1.82s/it]
                                                        

 15%|█▌        | 2270/14725 [1:11:49
+0: {'loss': 0.8827, 'grad_norm': 0.8205537215055217, 'learning_rate': 2e-05, 'epoch': 0.15}
+0: <6:16:53,  1.82s/it]
 15%|█▌        | 2271/14725 [1:11:51<6:17:50,  1.82s/it]
 15%|█▌        | 2272/14725 [1:11:53<6:18:14,  1.82s/it]
 15%|█▌        | 2273/14725 [1:11:55<6:21:13,  1.84s/it]
 15%|█▌        | 2274/14725 [1:11:56<6:16:09,  1.81s/it]
 15%|█▌        | 2275/14725 [1:11:58<6:15:19,  1.81s/it]
 15%|█▌        | 2276/14725 [1:12:00<6:15:36,  1.81s/it]
 15%|█▌        | 2277/14725 [1:12:02<6:16:26,  1.81s/it]
 15%|█▌        | 2278/14725 [1:12:04<6:12:59,  1.80s/it]
 15%|█▌        | 2279/14725 [1:12:05<6:15:55,  1.81s/it]
 15%|█▌        | 2280/14725 [1:12:07<6:18:12,  1.82s/it]
                                                        

 15%|█▌        | 2280/14725 [1:12:07<6:18:12,  1.82s/it]
 15%|█▌        | 2281/14725 [1:12:09<6:19:44,  1.83s/it]
 15%|█▌        | 2282/14725 [1:12:11<6:16:14,  1.81s/it]
 16%|█▌        | 2283/14725 [1:12:13<6:13:53,  1.80s/it]
 16%|█▌        | 2284/14725 [1:12:14<6:13:11,  1.80s/it]
 16%|█▌        | 2285/1472
+0: {'loss': 0.898, 'grad_norm': 0.8756028683089675, 'learning_rate': 2e-05, 'epoch': 0.16}
+0: 5 [1:12:16<6:16:55,  1.82s/it]
 16%|█▌        | 2286/14725 [1:12:18<6:20:15,  1.83s/it]
 16%|█▌        | 2287/14725 [1:12:20<6:18:22,  1.83s/it]
 16%|█▌        | 2288/14725 [1:12:22<6:16:12,  1.81s/it]
 16%|█▌        | 2289/14725 [1:12:24<6:18:02,  1.82s/it]
 16%|█▌        | 2290/14725 [1:12:25<6:17:05,  1.82s/it]
                                                        

 16%|█▌        | 2290/14725 [1:12:25<6:17:05,  1.82s/it]
 16%|█▌        | 2291/14725 [1:12:27<6:17:35,  1.82s/it]
 16%|█▌        | 2292/14725 [1:12:29<6:13:04,  1.80s/it]
 16%|█▌        | 2293/14725 [1:12:31<6:21:56,  1.84s/it]
 16%|█▌        | 2294/14725 [1:12:33<6:20:22,  1.84s/it]
 16%|█▌        | 2295/14725 [1:12:35<6:21:12,  1.84s/it]
 16%|█▌        | 2296/14725 [1:12:36<6:18:46,  1.83s/it]
 16%|█▌        | 2297/14725 [1:12:38<6:15:16,  1.81s/it]
 16%|█▌        | 2298/14725 [1:12:40<6:14:30,  1.81s/it]
 16%|█▌        | 2299/14725 [1:12:42<6:11:57,  1.80s/it]
 16%|█▌        |
+0: {'loss': 0.8455, 'grad_norm': 0.8005257142730992, 'learning_rate': 2e-05, 'epoch': 0.16}
+0: {'loss': 0.8646, 'grad_norm': 0.7678184182685271, 'learning_rate': 2e-05, 'epoch': 0.16}
+0:  2300/14725 [1:12:44<6:14:20,  1.81s/it]
                                                        

 16%|█▌        | 2300/14725 [1:12:44<6:14:20,  1.81s/it]
 16%|█▌        | 2301/14725 [1:12:45<6:16:43,  1.82s/it]
 16%|█▌        | 2302/14725 [1:12:47<6:15:57,  1.82s/it]
 16%|█▌        | 2303/14725 [1:12:49<6:16:35,  1.82s/it]
 16%|█▌        | 2304/14725 [1:12:51<6:19:34,  1.83s/it]
 16%|█▌        | 2305/14725 [1:12:53<6:16:14,  1.82s/it]
 16%|█▌        | 2306/14725 [1:12:54<6:15:23,  1.81s/it]
 16%|█▌        | 2307/14725 [1:12:56<6:21:52,  1.85s/it]
 16%|█▌        | 2308/14725 [1:12:58<6:19:58,  1.84s/it]
 16%|█▌        | 2309/14725 [1:13:00<6:24:00,  1.86s/it]
 16%|█▌        | 2310/14725 [1:13:02<6:21:58,  1.85s/it]
                                                        

 16%|█▌        | 2310/14725 [1:13:02<6:21:58,  1.85s/it]
 16%|█▌        | 2311/14725 [1:13:04<6:23:09,  1.85s/it]
 16%|█▌        | 2312/14725 [1:13:06<6:28:19,  1.88s/it]
 16%|█▌  
+0: {'loss': 0.8741, 'grad_norm': 0.7819529376910934, 'learning_rate': 2e-05, 'epoch': 0.16}
+0:       | 2313/14725 [1:13:08<6:24:24,  1.86s/it]
 16%|█▌        | 2314/14725 [1:13:09<6:21:20,  1.84s/it]
 16%|█▌        | 2315/14725 [1:13:11<6:22:08,  1.85s/it]
 16%|█▌        | 2316/14725 [1:13:13<6:18:19,  1.83s/it]
 16%|█▌        | 2317/14725 [1:13:15<6:18:33,  1.83s/it]
 16%|█▌        | 2318/14725 [1:13:17<6:14:52,  1.81s/it]
 16%|█▌        | 2319/14725 [1:13:18<6:11:17,  1.80s/it]
 16%|█▌        | 2320/14725 [1:13:20<6:10:37,  1.79s/it]
                                                        

 16%|█▌        | 2320/14725 [1:13:20<6:10:37,  1.79s/it]
 16%|█▌        | 2321/14725 [1:13:22<6:09:59,  1.79s/it]
 16%|█▌        | 2322/14725 [1:13:24<6:13:16,  1.81s/it]
 16%|█▌        | 2323/14725 [1:13:26<6:13:10,  1.81s/it]
 16%|█▌        | 2324/14725 [1:13:27<6:16:08,  1.82s/it]
 16%|█▌        | 2325/14725 [1:13:29<6:22:11,  1.85s/it]
 16%|█▌        | 2326/14725 [1:13:31<6:19:42,  1.84s/it]
 16%|█▌        | 2327/14725 [1:13:33<6:15:57,  1.82s/it]
 16
+0: {'loss': 0.8934, 'grad_norm': 0.834369612752762, 'learning_rate': 2e-05, 'epoch': 0.16}
+0: {'loss': 0.8689, 'grad_norm': 0.8637130699934877, 'learning_rate': 2e-05, 'epoch': 0.16}
+0: %|█▌        | 2328/14725 [1:13:35<6:12:05,  1.80s/it]
 16%|█▌        | 2329/14725 [1:13:36<6:10:16,  1.79s/it]
 16%|█▌        | 2330/14725 [1:13:38<6:09:26,  1.79s/it]
                                                        

 16%|█▌        | 2330/14725 [1:13:38<6:09:26,  1.79s/it]
 16%|█▌        | 2331/14725 [1:13:40<6:11:07,  1.80s/it]
 16%|█▌        | 2332/14725 [1:13:42<6:12:50,  1.81s/it]
 16%|█▌        | 2333/14725 [1:13:44<6:12:52,  1.81s/it]
 16%|█▌        | 2334/14725 [1:13:46<6:11:44,  1.80s/it]
 16%|█▌        | 2335/14725 [1:13:47<6:14:10,  1.81s/it]
 16%|█▌        | 2336/14725 [1:13:49<6:11:00,  1.80s/it]
 16%|█▌        | 2337/14725 [1:13:51<6:14:42,  1.81s/it]
 16%|█▌        | 2338/14725 [1:13:53<6:12:09,  1.80s/it]
 16%|█▌        | 2339/14725 [1:13:55<6:21:58,  1.85s/it]
 16%|█▌        | 2340/14725 [1:13:57<6:22:27,  1.85s/it]
                                                        

 16%|█▌        | 2340/14725 [1:13:57<6:22:27,  1.85s/
+0: {'loss': 0.8932, 'grad_norm': 0.8038221967035035, 'learning_rate': 2e-05, 'epoch': 0.16}
+0: it]
 16%|█▌        | 2341/14725 [1:13:58<6:18:00,  1.83s/it]
 16%|█▌        | 2342/14725 [1:14:00<6:18:07,  1.83s/it]
 16%|█▌        | 2343/14725 [1:14:02<6:18:34,  1.83s/it]
 16%|█▌        | 2344/14725 [1:14:04<6:15:34,  1.82s/it]
 16%|█▌        | 2345/14725 [1:14:06<6:21:34,  1.85s/it]
 16%|█▌        | 2346/14725 [1:14:07<6:16:03,  1.82s/it]
 16%|█▌        | 2347/14725 [1:14:09<6:18:01,  1.83s/it]
 16%|█▌        | 2348/14725 [1:14:11<6:14:56,  1.82s/it]
 16%|█▌        | 2349/14725 [1:14:13<6:12:20,  1.81s/it]
 16%|█▌        | 2350/14725 [1:14:15<6:09:26,  1.79s/it]
                                                        

 16%|█▌        | 2350/14725 [1:14:15<6:09:26,  1.79s/it]
 16%|█▌        | 2351/14725 [1:14:16<6:11:45,  1.80s/it]
 16%|█▌        | 2352/14725 [1:14:18<6:08:37,  1.79s/it]
 16%|█▌        | 2353/14725 [1:14:20<6:10:44,  1.80s/it]
 16%|█▌        | 2354/14725 [1:14:22<6:09:09,  1.79s/it]
 16%|█▌        | 2355/14725 [1:14:24<6:07:2
+0: {'loss': 0.8678, 'grad_norm': 0.8159595802068723, 'learning_rate': 2e-05, 'epoch': 0.16}
+0: 6,  1.78s/it]
 16%|█▌        | 2356/14725 [1:14:26<6:17:12,  1.83s/it]
 16%|█▌        | 2357/14725 [1:14:27<6:11:23,  1.80s/it]
 16%|█▌        | 2358/14725 [1:14:29<6:11:42,  1.80s/it]
 16%|█▌        | 2359/14725 [1:14:31<6:15:51,  1.82s/it]
 16%|█▌        | 2360/14725 [1:14:33<6:15:56,  1.82s/it]
                                                        

 16%|█▌        | 2360/14725 [1:14:33<6:15:56,  1.82s/it]
 16%|█▌        | 2361/14725 [1:14:35<6:15:39,  1.82s/it]
 16%|█▌        | 2362/14725 [1:14:36<6:13:55,  1.81s/it]
 16%|█▌        | 2363/14725 [1:14:38<6:12:18,  1.81s/it]
 16%|█▌        | 2364/14725 [1:14:40<6:17:30,  1.83s/it]
 16%|█▌        | 2365/14725 [1:14:42<6:20:25,  1.85s/it]
 16%|█▌        | 2366/14725 [1:14:44<6:19:02,  1.84s/it]
 16%|█▌        | 2367/14725 [1:14:46<6:14:32,  1.82s/it]
 16%|█▌        | 2368/14725 [1:14:47<6:17:03,  1.83s/it]
 16%|█▌        | 2369/14725 [1:14:49<6:13:03,  1.81s/it]
 16%|█▌        | 2370/14725 [1:14
+0: {'loss': 0.8835, 'grad_norm': 0.7777923561946707, 'learning_rate': 2e-05, 'epoch': 0.16}
+0: {'loss': 0.8571, 'grad_norm': 0.857238639808066, 'learning_rate': 2e-05, 'epoch': 0.16}
+0: :51<6:12:51,  1.81s/it]
                                                        

 16%|█▌        | 2370/14725 [1:14:51<6:12:51,  1.81s/it]
 16%|█▌        | 2371/14725 [1:14:53<6:20:43,  1.85s/it]
 16%|█▌        | 2372/14725 [1:14:55<6:19:01,  1.84s/it]
 16%|█▌        | 2373/14725 [1:14:57<6:16:41,  1.83s/it]
 16%|█▌        | 2374/14725 [1:14:58<6:13:16,  1.81s/it]
 16%|█▌        | 2375/14725 [1:15:00<6:16:22,  1.83s/it]
 16%|█▌        | 2376/14725 [1:15:02<6:13:23,  1.81s/it]
 16%|█▌        | 2377/14725 [1:15:04<6:17:32,  1.83s/it]
 16%|█▌        | 2378/14725 [1:15:06<6:26:50,  1.88s/it]
 16%|█▌        | 2379/14725 [1:15:08<6:20:48,  1.85s/it]
 16%|█▌        | 2380/14725 [1:15:09<6:18:33,  1.84s/it]
                                                        

 16%|█▌        | 2380/14725 [1:15:09<6:18:33,  1.84s/it]
 16%|█▌        | 2381/14725 [1:15:11<6:20:53,  1.85s/it]
 16%|█▌        | 2382/14725 [1:15:13<6:20:41,  1.85s/it]
 16%|█▌        | 2383/1472
+0: {'loss': 0.8829, 'grad_norm': 0.8535039674231056, 'learning_rate': 2e-05, 'epoch': 0.16}
+0: 5 [1:15:15<6:16:13,  1.83s/it]
 16%|█▌        | 2384/14725 [1:15:17<6:12:04,  1.81s/it]
 16%|█▌        | 2385/14725 [1:15:18<6:08:39,  1.79s/it]
 16%|█▌        | 2386/14725 [1:15:20<6:10:48,  1.80s/it]
 16%|█▌        | 2387/14725 [1:15:22<6:11:12,  1.81s/it]
 16%|█▌        | 2388/14725 [1:15:24<6:09:01,  1.79s/it]
 16%|█▌        | 2389/14725 [1:15:26<6:09:31,  1.80s/it]
 16%|█▌        | 2390/14725 [1:15:28<6:12:29,  1.81s/it]
                                                        

 16%|█▌        | 2390/14725 [1:15:28<6:12:29,  1.81s/it]
 16%|█▌        | 2391/14725 [1:15:29<6:11:43,  1.81s/it]
 16%|█▌        | 2392/14725 [1:15:31<6:13:17,  1.82s/it]
 16%|█▋        | 2393/14725 [1:15:33<6:15:03,  1.82s/it]
 16%|█▋        | 2394/14725 [1:15:35<6:11:50,  1.81s/it]
 16%|█▋        | 2395/14725 [1:15:37<6:12:44,  1.81s/it]
 16%|█▋        | 2396/14725 [1:15:38<6:10:45,  1.80s/it]
 16%|█▋        | 2397/14725 [1:15:40<6:13:09,  1.82s/it]
 16%|█▋        |
+0: {'loss': 0.8874, 'grad_norm': 0.8419011610447622, 'learning_rate': 2e-05, 'epoch': 0.16}
+0: {'loss': 0.8671, 'grad_norm': 0.8613424265757127, 'learning_rate': 2e-05, 'epoch': 0.16}
+0:  2398/14725 [1:15:42<6:12:15,  1.81s/it]
 16%|█▋        | 2399/14725 [1:15:44<6:17:45,  1.84s/it]
 16%|█▋        | 2400/14725 [1:15:46<6:17:01,  1.84s/it]
                                                        

 16%|█▋        | 2400/14725 [1:15:46<6:17:01,  1.84s/it]
 16%|█▋        | 2401/14725 [1:15:48<6:15:50,  1.83s/it]
 16%|█▋        | 2402/14725 [1:15:49<6:18:40,  1.84s/it]
 16%|█▋        | 2403/14725 [1:15:51<6:14:24,  1.82s/it]
 16%|█▋        | 2404/14725 [1:15:53<6:11:46,  1.81s/it]
 16%|█▋        | 2405/14725 [1:15:55<6:09:27,  1.80s/it]
 16%|█▋        | 2406/14725 [1:15:57<6:10:11,  1.80s/it]
 16%|█▋        | 2407/14725 [1:15:58<6:08:34,  1.80s/it]
 16%|█▋        | 2408/14725 [1:16:00<6:07:51,  1.79s/it]
 16%|█▋        | 2409/14725 [1:16:02<6:10:15,  1.80s/it]
 16%|█▋        | 2410/14725 [1:16:04<6:07:23,  1.79s/it]
                                                        

 16%|█▋        | 2410/14725 [1:16:04<6:07:23,  1.79s/it]
 16%|█▋  
+0: {'loss': 0.8753, 'grad_norm': 0.844241952945411, 'learning_rate': 2e-05, 'epoch': 0.16}
+0:       | 2411/14725 [1:16:06<6:09:44,  1.80s/it]
 16%|█▋        | 2412/14725 [1:16:07<6:13:07,  1.82s/it]
 16%|█▋        | 2413/14725 [1:16:09<6:09:54,  1.80s/it]
 16%|█▋        | 2414/14725 [1:16:11<6:10:55,  1.81s/it]
 16%|█▋        | 2415/14725 [1:16:13<6:09:23,  1.80s/it]
 16%|█▋        | 2416/14725 [1:16:15<6:06:51,  1.79s/it]
 16%|█▋        | 2417/14725 [1:16:16<6:09:10,  1.80s/it]
 16%|█▋        | 2418/14725 [1:16:18<6:07:56,  1.79s/it]
 16%|█▋        | 2419/14725 [1:16:20<6:10:27,  1.81s/it]
 16%|█▋        | 2420/14725 [1:16:22<6:07:48,  1.79s/it]
                                                        

 16%|█▋        | 2420/14725 [1:16:22<6:07:48,  1.79s/it]
 16%|█▋        | 2421/14725 [1:16:24<6:10:26,  1.81s/it]
 16%|█▋        | 2422/14725 [1:16:25<6:10:36,  1.81s/it]
 16%|█▋        | 2423/14725 [1:16:27<6:07:17,  1.79s/it]
 16%|█▋        | 2424/14725 [1:16:29<6:08:21,  1.80s/it]
 16%|█▋        | 2425/14725 [1:16:31<6:06:19,  1.79s/it]
 16
+0: {'loss': 0.8619, 'grad_norm': 0.8224695841414393, 'learning_rate': 2e-05, 'epoch': 0.17}
+0: %|█▋        | 2426/14725 [1:16:32<6:06:04,  1.79s/it]
 16%|█▋        | 2427/14725 [1:16:34<6:04:45,  1.78s/it]
 16%|█▋        | 2428/14725 [1:16:36<6:04:04,  1.78s/it]
 16%|█▋        | 2429/14725 [1:16:38<6:10:57,  1.81s/it]
 17%|█▋        | 2430/14725 [1:16:40<6:08:32,  1.80s/it]
                                                        

 17%|█▋        | 2430/14725 [1:16:40<6:08:32,  1.80s/it]
 17%|█▋        | 2431/14725 [1:16:41<6:08:26,  1.80s/it]
 17%|█▋        | 2432/14725 [1:16:43<6:05:49,  1.79s/it]
 17%|█▋        | 2433/14725 [1:16:45<6:13:56,  1.83s/it]
 17%|█▋        | 2434/14725 [1:16:47<6:12:33,  1.82s/it]
 17%|█▋        | 2435/14725 [1:16:49<6:08:51,  1.80s/it]
 17%|█▋        | 2436/14725 [1:16:51<6:16:50,  1.84s/it]
 17%|█▋        | 2437/14725 [1:16:53<6:17:34,  1.84s/it]
 17%|█▋        | 2438/14725 [1:16:54<6:13:13,  1.82s/it]
 17%|█▋        | 2439/14725 [1:16:56<6:14:07,  1.83s/it]
 17%|█▋        | 2440/14725 [1:16:58<6:29:31,  1.9
+0: {'loss': 0.8789, 'grad_norm': 0.8519918667507355, 'learning_rate': 2e-05, 'epoch': 0.17}
+0: {'loss': 0.8675, 'grad_norm': 0.771685373086738, 'learning_rate': 2e-05, 'epoch': 0.17}
+0: 0s/it]
                                                        

 17%|█▋        | 2440/14725 [1:16:58<6:29:31,  1.90s/it]
 17%|█▋        | 2441/14725 [1:17:00<6:23:29,  1.87s/it]
 17%|█▋        | 2442/14725 [1:17:02<6:16:55,  1.84s/it]
 17%|█▋        | 2443/14725 [1:17:04<6:12:07,  1.82s/it]
 17%|█▋        | 2444/14725 [1:17:05<6:09:15,  1.80s/it]
 17%|█▋        | 2445/14725 [1:17:07<6:07:26,  1.80s/it]
 17%|█▋        | 2446/14725 [1:17:09<6:07:50,  1.80s/it]
 17%|█▋        | 2447/14725 [1:17:11<6:04:44,  1.78s/it]
 17%|█▋        | 2448/14725 [1:17:12<6:05:03,  1.78s/it]
 17%|█▋        | 2449/14725 [1:17:14<6:06:33,  1.79s/it]
 17%|█▋        | 2450/14725 [1:17:16<6:16:52,  1.84s/it]
                                                        

 17%|█▋        | 2450/14725 [1:17:16<6:16:52,  1.84s/it]
 17%|█▋        | 2451/14725 [1:17:18<6:15:08,  1.83s/it]
 17%|█▋        | 2452/14725 [1:17:20<6:22:20,  1.87s/it]
 17%|█▋        | 2453/14725 [1:17:22<6:20:5
+0: {'loss': 0.8959, 'grad_norm': 0.8126132346418734, 'learning_rate': 2e-05, 'epoch': 0.17}
+0: 8,  1.86s/it]
 17%|█▋        | 2454/14725 [1:17:24<6:16:40,  1.84s/it]
 17%|█▋        | 2455/14725 [1:17:25<6:14:54,  1.83s/it]
 17%|█▋        | 2456/14725 [1:17:27<6:19:39,  1.86s/it]
 17%|█▋        | 2457/14725 [1:17:29<6:21:01,  1.86s/it]
 17%|█▋        | 2458/14725 [1:17:31<6:18:08,  1.85s/it]
 17%|█▋        | 2459/14725 [1:17:33<6:17:49,  1.85s/it]
 17%|█▋        | 2460/14725 [1:17:35<6:17:02,  1.84s/it]
                                                        

 17%|█▋        | 2460/14725 [1:17:35<6:17:02,  1.84s/it]
 17%|█▋        | 2461/14725 [1:17:37<6:19:35,  1.86s/it]
 17%|█▋        | 2462/14725 [1:17:38<6:16:54,  1.84s/it]
 17%|█▋        | 2463/14725 [1:17:40<6:14:05,  1.83s/it]
 17%|█▋        | 2464/14725 [1:17:42<6:15:43,  1.84s/it]
 17%|█▋        | 2465/14725 [1:17:44<6:13:03,  1.83s/it]
 17%|█▋        | 2466/14725 [1:17:46<6:12:58,  1.83s/it]
 17%|█▋        | 2467/14725 [1:17:47<6:11:51,  1.82s/it]
 17%|█▋        | 2468/14725 [1:17
+0: {'loss': 0.8839, 'grad_norm': 0.8137389106899859, 'learning_rate': 2e-05, 'epoch': 0.17}
+0: {'loss': 0.8602, 'grad_norm': 0.7839510089014174, 'learning_rate': 2e-05, 'epoch': 0.17}
+0: :49<6:08:53,  1.81s/it]
 17%|█▋        | 2469/14725 [1:17:51<6:09:07,  1.81s/it]
 17%|█▋        | 2470/14725 [1:17:53<6:11:31,  1.82s/it]
                                                        

 17%|█▋        | 2470/14725 [1:17:53<6:11:31,  1.82s/it]
 17%|█▋        | 2471/14725 [1:17:55<6:12:38,  1.82s/it]
 17%|█▋        | 2472/14725 [1:17:57<6:10:20,  1.81s/it]
 17%|█▋        | 2473/14725 [1:17:58<6:07:54,  1.80s/it]
 17%|█▋        | 2474/14725 [1:18:00<6:07:33,  1.80s/it]
 17%|█▋        | 2475/14725 [1:18:02<6:11:24,  1.82s/it]
 17%|█▋        | 2476/14725 [1:18:04<6:12:04,  1.82s/it]
 17%|█▋        | 2477/14725 [1:18:06<6:06:46,  1.80s/it]
 17%|█▋        | 2478/14725 [1:18:07<6:04:54,  1.79s/it]
 17%|█▋        | 2479/14725 [1:18:09<6:03:54,  1.78s/it]
 17%|█▋        | 2480/14725 [1:18:11<6:04:23,  1.79s/it]
                                                        

 17%|█▋        | 2480/14725 [1:18:11<6:04:23,  1.79s/it]
 17%|█▋        | 2481/1472
+0: {'loss': 0.864, 'grad_norm': 0.8345181258627244, 'learning_rate': 2e-05, 'epoch': 0.17}
+0: 5 [1:18:13<6:03:29,  1.78s/it]
 17%|█▋        | 2482/14725 [1:18:14<6:02:19,  1.78s/it]
 17%|█▋        | 2483/14725 [1:18:16<6:08:04,  1.80s/it]
 17%|█▋        | 2484/14725 [1:18:18<6:06:52,  1.80s/it]
 17%|█▋        | 2485/14725 [1:18:20<6:11:25,  1.82s/it]
 17%|█▋        | 2486/14725 [1:18:22<6:09:47,  1.81s/it]
 17%|█▋        | 2487/14725 [1:18:24<6:13:25,  1.83s/it]
 17%|█▋        | 2488/14725 [1:18:26<6:20:48,  1.87s/it]
 17%|█▋        | 2489/14725 [1:18:27<6:14:41,  1.84s/it]
 17%|█▋        | 2490/14725 [1:18:29<6:11:49,  1.82s/it]
                                                        

 17%|█▋        | 2490/14725 [1:18:29<6:11:49,  1.82s/it]
 17%|█▋        | 2491/14725 [1:18:31<6:10:53,  1.82s/it]
 17%|█▋        | 2492/14725 [1:18:33<6:09:29,  1.81s/it]
 17%|█▋        | 2493/14725 [1:18:35<6:11:51,  1.82s/it]
 17%|█▋        | 2494/14725 [1:18:36<6:16:58,  1.85s/it]
 17%|█▋        | 2495/14725 [1:18:38<6:13:21,  1.83s/it]
 17%|█▋        |
+0: {'loss': 0.8728, 'grad_norm': 0.7977809424296695, 'learning_rate': 2e-05, 'epoch': 0.17}
+0:  2496/14725 [1:18:40<6:10:45,  1.82s/it]
 17%|█▋        | 2497/14725 [1:18:42<6:11:44,  1.82s/it]
 17%|█▋        | 2498/14725 [1:18:44<6:11:51,  1.82s/it]
 17%|█▋        | 2499/14725 [1:18:45<6:08:32,  1.81s/it]
 17%|█▋        | 2500/14725 [1:18:47<6:10:09,  1.82s/it]
                                                        

 17%|█▋        | 2500/14725 [1:18:47<6:10:09,  1.82s/it]
 17%|█▋        | 2501/14725 [1:18:49<6:08:09,  1.81s/it]
 17%|█▋        | 2502/14725 [1:18:51<6:05:59,  1.80s/it]
 17%|█▋        | 2503/14725 [1:18:53<6:10:00,  1.82s/it]
 17%|█▋        | 2504/14725 [1:18:55<6:09:45,  1.82s/it]
 17%|█▋        | 2505/14725 [1:18:56<6:10:46,  1.82s/it]
 17%|█▋        | 2506/14725 [1:18:58<6:17:15,  1.85s/it]
 17%|█▋        | 2507/14725 [1:19:00<6:13:26,  1.83s/it]
 17%|█▋        | 2508/14725 [1:19:02<6:10:01,  1.82s/it]
 17%|█▋        | 2509/14725 [1:19:04<6:10:43,  1.82s/it]
 17%|█▋        | 2510/14725 [1:19:06<6:09:40,  1.82s/it]
          
+0: {'loss': 0.8634, 'grad_norm': 0.7853955582837857, 'learning_rate': 2e-05, 'epoch': 0.17}
+0: {'loss': 0.8894, 'grad_norm': 0.7814718930604274, 'learning_rate': 2e-05, 'epoch': 0.17}
+0:                                               

 17%|█▋        | 2510/14725 [1:19:06<6:09:40,  1.82s/it]
 17%|█▋        | 2511/14725 [1:19:07<6:13:44,  1.84s/it]
 17%|█▋        | 2512/14725 [1:19:09<6:11:01,  1.82s/it]
 17%|█▋        | 2513/14725 [1:19:11<6:05:53,  1.80s/it]
 17%|█▋        | 2514/14725 [1:19:13<6:05:12,  1.79s/it]
 17%|█▋        | 2515/14725 [1:19:15<6:10:04,  1.82s/it]
 17%|█▋        | 2516/14725 [1:19:16<6:11:37,  1.83s/it]
 17%|█▋        | 2517/14725 [1:19:18<6:07:53,  1.81s/it]
 17%|█▋        | 2518/14725 [1:19:20<6:09:18,  1.82s/it]
 17%|█▋        | 2519/14725 [1:19:22<6:10:07,  1.82s/it]
 17%|█▋        | 2520/14725 [1:19:24<6:07:37,  1.81s/it]
                                                        

 17%|█▋        | 2520/14725 [1:19:24<6:07:37,  1.81s/it]
 17%|█▋        | 2521/14725 [1:19:25<6:08:05,  1.81s/it]
 17%|█▋        | 2522/14725 [1:19:27<6:08:51,  1.81s/it]
 17%|█▋        | 2523/14725 [1:19:29<6:14:12,  1.84s/it]
 17
+0: {'loss': 0.8643, 'grad_norm': 0.8428264675531854, 'learning_rate': 2e-05, 'epoch': 0.17}
+0: %|█▋        | 2524/14725 [1:19:31<6:17:00,  1.85s/it]
 17%|█▋        | 2525/14725 [1:19:33<6:12:43,  1.83s/it]
 17%|█▋        | 2526/14725 [1:19:35<6:16:10,  1.85s/it]
 17%|█▋        | 2527/14725 [1:19:37<6:11:45,  1.83s/it]
 17%|█▋        | 2528/14725 [1:19:38<6:07:00,  1.81s/it]
 17%|█▋        | 2529/14725 [1:19:40<6:06:54,  1.81s/it]
 17%|█▋        | 2530/14725 [1:19:42<6:09:04,  1.82s/it]
                                                        

 17%|█▋        | 2530/14725 [1:19:42<6:09:04,  1.82s/it]
 17%|█▋        | 2531/14725 [1:19:44<6:06:29,  1.80s/it]
 17%|█▋        | 2532/14725 [1:19:45<6:04:19,  1.79s/it]
 17%|█▋        | 2533/14725 [1:19:47<6:02:15,  1.78s/it]
 17%|█▋        | 2534/14725 [1:19:49<6:03:46,  1.79s/it]
 17%|█▋        | 2535/14725 [1:19:51<6:05:15,  1.80s/it]
 17%|█▋        | 2536/14725 [1:19:53<6:08:29,  1.81s/it]
 17%|█▋        | 2537/14725 [1:19:54<6:06:35,  1.80s/it]
 17%|█▋        | 2538/14725 [1:19:56<6:05:26,  1.8
+0: {'loss': 0.8749, 'grad_norm': 0.8165450772744813, 'learning_rate': 2e-05, 'epoch': 0.17}
+0: {'loss': 0.8709, 'grad_norm': 0.809634068336782, 'learning_rate': 2e-05, 'epoch': 0.17}
+0: 0s/it]
 17%|█▋        | 2539/14725 [1:19:58<6:02:03,  1.78s/it]
 17%|█▋        | 2540/14725 [1:20:00<6:03:25,  1.79s/it]
                                                        

 17%|█▋        | 2540/14725 [1:20:00<6:03:25,  1.79s/it]
 17%|█▋        | 2541/14725 [1:20:02<6:03:45,  1.79s/it]
 17%|█▋        | 2542/14725 [1:20:03<6:05:01,  1.80s/it]
 17%|█▋        | 2543/14725 [1:20:05<6:03:31,  1.79s/it]
 17%|█▋        | 2544/14725 [1:20:07<6:10:08,  1.82s/it]
 17%|█▋        | 2545/14725 [1:20:09<6:06:18,  1.80s/it]
 17%|█▋        | 2546/14725 [1:20:11<6:09:54,  1.82s/it]
 17%|█▋        | 2547/14725 [1:20:13<6:07:29,  1.81s/it]
 17%|█▋        | 2548/14725 [1:20:14<6:05:29,  1.80s/it]
 17%|█▋        | 2549/14725 [1:20:16<6:11:03,  1.83s/it]
 17%|█▋        | 2550/14725 [1:20:18<6:08:23,  1.82s/it]
                                                        

 17%|█▋        | 2550/14725 [1:20:18<6:08:23,  1.82s/it]
 17%|█▋        | 2551/14725 [1:20:20<6:07:2
+0: {'loss': 0.8852, 'grad_norm': 0.8428719885646948, 'learning_rate': 2e-05, 'epoch': 0.17}
+0: 4,  1.81s/it]
 17%|█▋        | 2552/14725 [1:20:22<6:03:58,  1.79s/it]
 17%|█▋        | 2553/14725 [1:20:23<6:05:26,  1.80s/it]
 17%|█▋        | 2554/14725 [1:20:25<6:03:15,  1.79s/it]
 17%|█▋        | 2555/14725 [1:20:27<6:03:27,  1.79s/it]
 17%|█▋        | 2556/14725 [1:20:29<6:26:54,  1.91s/it]
 17%|█▋        | 2557/14725 [1:20:31<6:20:33,  1.88s/it]
 17%|█▋        | 2558/14725 [1:20:33<6:49:37,  2.02s/it]
 17%|█▋        | 2559/14725 [1:20:35<6:41:09,  1.98s/it]
 17%|█▋        | 2560/14725 [1:20:37<6:31:06,  1.93s/it]
                                                        

 17%|█▋        | 2560/14725 [1:20:37<6:31:06,  1.93s/it]
 17%|█▋        | 2561/14725 [1:20:39<6:21:44,  1.88s/it]
 17%|█▋        | 2562/14725 [1:20:40<6:15:35,  1.85s/it]
 17%|█▋        | 2563/14725 [1:20:42<6:14:04,  1.85s/it]
 17%|█▋        | 2564/14725 [1:20:44<6:13:12,  1.84s/it]
 17%|█▋        | 2565/14725 [1:20:46<6:09:48,  1.82s/it]
 17%|█▋        | 2566/14725 [1:20
+0: {'loss': 0.8858, 'grad_norm': 0.8204073048733921, 'learning_rate': 2e-05, 'epoch': 0.17}
+0: :48<6:35:32,  1.95s/it]
 17%|█▋        | 2567/14725 [1:20:50<6:26:59,  1.91s/it]
 17%|█▋        | 2568/14725 [1:20:52<6:25:11,  1.90s/it]
 17%|█▋        | 2569/14725 [1:20:54<6:20:09,  1.88s/it]
 17%|█▋        | 2570/14725 [1:20:56<6:17:39,  1.86s/it]
                                                        

 17%|█▋        | 2570/14725 [1:20:56<6:17:39,  1.86s/it]
 17%|█▋        | 2571/14725 [1:20:57<6:16:06,  1.86s/it]
 17%|█▋        | 2572/14725 [1:20:59<6:10:51,  1.83s/it]
 17%|█▋        | 2573/14725 [1:21:01<6:11:15,  1.83s/it]
 17%|█▋        | 2574/14725 [1:21:03<6:09:08,  1.82s/it]
 17%|█▋        | 2575/14725 [1:21:05<6:05:56,  1.81s/it]
 17%|█▋        | 2576/14725 [1:21:07<6:34:15,  1.95s/it]
 18%|█▊        | 2577/14725 [1:21:09<6:27:31,  1.91s/it]
 18%|█▊        | 2578/14725 [1:21:11<6:31:39,  1.93s/it]
 18%|█▊        | 2579/14725 [1:21:12<6:24:24,  1.90s/it]
 18%|█▊        | 2580/14725 [1:21:14<6:21:39,  1.89s/it]
                           
+0: {'loss': 0.879, 'grad_norm': 0.8370715725795197, 'learning_rate': 2e-05, 'epoch': 0.18}
+0: {'loss': 0.8716, 'grad_norm': 0.7715128853512155, 'learning_rate': 2e-05, 'epoch': 0.18}
+0:                              

 18%|█▊        | 2580/14725 [1:21:14<6:21:39,  1.89s/it]
 18%|█▊        | 2581/14725 [1:21:16<6:16:08,  1.86s/it]
 18%|█▊        | 2582/14725 [1:21:18<6:10:26,  1.83s/it]
 18%|█▊        | 2583/14725 [1:21:20<6:06:17,  1.81s/it]
 18%|█▊        | 2584/14725 [1:21:21<6:06:53,  1.81s/it]
 18%|█▊        | 2585/14725 [1:21:23<6:09:24,  1.83s/it]
 18%|█▊        | 2586/14725 [1:21:25<6:09:04,  1.82s/it]
 18%|█▊        | 2587/14725 [1:21:27<6:17:54,  1.87s/it]
 18%|█▊        | 2588/14725 [1:21:29<6:14:39,  1.85s/it]
 18%|█▊        | 2589/14725 [1:21:31<6:17:23,  1.87s/it]
 18%|█▊        | 2590/14725 [1:21:33<6:11:47,  1.84s/it]
                                                        

 18%|█▊        | 2590/14725 [1:21:33<6:11:47,  1.84s/it]
 18%|█▊        | 2591/14725 [1:21:34<6:12:51,  1.84s/it]
 18%|█▊        | 2592/14725 [1:21:36<6:13:25,  1.85s/it]
 18%|█▊        | 2593/14725 [1:21:38<6:11:46,  1.84s/it]
 18%|█▊        |
+0: {'loss': 0.8773, 'grad_norm': 0.8230177631039766, 'learning_rate': 2e-05, 'epoch': 0.18}
+0:  2594/14725 [1:21:40<6:08:31,  1.82s/it]
 18%|█▊        | 2595/14725 [1:21:42<6:08:18,  1.82s/it]
 18%|█▊        | 2596/14725 [1:21:44<6:06:38,  1.81s/it]
 18%|█▊        | 2597/14725 [1:21:45<6:10:13,  1.83s/it]
 18%|█▊        | 2598/14725 [1:21:47<6:06:20,  1.81s/it]
 18%|█▊        | 2599/14725 [1:21:49<6:09:05,  1.83s/it]
 18%|█▊        | 2600/14725 [1:21:51<6:08:44,  1.82s/it]
                                                        

 18%|█▊        | 2600/14725 [1:21:51<6:08:44,  1.82s/it]
 18%|█▊        | 2601/14725 [1:21:53<6:09:13,  1.83s/it]
 18%|█▊        | 2602/14725 [1:21:54<6:07:07,  1.82s/it]
 18%|█▊        | 2603/14725 [1:21:56<6:02:55,  1.80s/it]
 18%|█▊        | 2604/14725 [1:21:58<6:07:49,  1.82s/it]
 18%|█▊        | 2605/14725 [1:22:00<6:05:02,  1.81s/it]
 18%|█▊        | 2606/14725 [1:22:02<6:08:32,  1.82s/it]
 18%|█▊        | 2607/14725 [1:22:04<6:06:01,  1.81s/it]
 18%|█▊        | 2608/14725 [1:22:05<6:02:23,  1.79s/it]
 18%|█�
+0: {'loss': 0.8608, 'grad_norm': 0.85617189469023, 'learning_rate': 2e-05, 'epoch': 0.18}
+0: {'loss': 0.8551, 'grad_norm': 0.7825434640630082, 'learning_rate': 2e-05, 'epoch': 0.18}
+0: �        | 2609/14725 [1:22:07<6:01:09,  1.79s/it]
 18%|█▊        | 2610/14725 [1:22:09<6:00:40,  1.79s/it]
                                                        

 18%|█▊        | 2610/14725 [1:22:09<6:00:40,  1.79s/it]
 18%|█▊        | 2611/14725 [1:22:11<6:00:18,  1.78s/it]
 18%|█▊        | 2612/14725 [1:22:12<6:07:06,  1.82s/it]
 18%|█▊        | 2613/14725 [1:22:14<6:04:06,  1.80s/it]
 18%|█▊        | 2614/14725 [1:22:16<6:02:40,  1.80s/it]
 18%|█▊        | 2615/14725 [1:22:18<6:02:57,  1.80s/it]
 18%|█▊        | 2616/14725 [1:22:20<6:03:12,  1.80s/it]
 18%|█▊        | 2617/14725 [1:22:22<6:08:14,  1.82s/it]
 18%|█▊        | 2618/14725 [1:22:23<6:11:29,  1.84s/it]
 18%|█▊        | 2619/14725 [1:22:25<6:10:48,  1.84s/it]
 18%|█▊        | 2620/14725 [1:22:27<6:09:31,  1.83s/it]
                                                        

 18%|█▊        | 2620/14725 [1:22:27<6:09:31,  1.83s/it]
 18%|█▊        | 2621/14725 [1:22:29<6:09:46,  1.83s/it]
 18
+0: {'loss': 0.8702, 'grad_norm': 0.789939156621892, 'learning_rate': 2e-05, 'epoch': 0.18}
+0: %|█▊        | 2622/14725 [1:22:31<6:11:45,  1.84s/it]
 18%|█▊        | 2623/14725 [1:22:33<6:11:50,  1.84s/it]
 18%|█▊        | 2624/14725 [1:22:34<6:13:31,  1.85s/it]
 18%|█▊        | 2625/14725 [1:22:36<6:10:23,  1.84s/it]
 18%|█▊        | 2626/14725 [1:22:38<6:07:36,  1.82s/it]
 18%|█▊        | 2627/14725 [1:22:40<6:04:58,  1.81s/it]
 18%|█▊        | 2628/14725 [1:22:42<6:01:23,  1.79s/it]
 18%|█▊        | 2629/14725 [1:22:43<6:00:03,  1.79s/it]
 18%|█▊        | 2630/14725 [1:22:45<6:02:07,  1.80s/it]
                                                        

 18%|█▊        | 2630/14725 [1:22:45<6:02:07,  1.80s/it]
 18%|█▊        | 2631/14725 [1:22:47<6:02:29,  1.80s/it]
 18%|█▊        | 2632/14725 [1:22:49<6:14:47,  1.86s/it]
 18%|█▊        | 2633/14725 [1:22:51<6:10:50,  1.84s/it]
 18%|█▊        | 2634/14725 [1:22:53<6:05:39,  1.81s/it]
 18%|█▊        | 2635/14725 [1:22:54<6:10:34,  1.84s/it]
 18%|█▊        | 2636/14725 [1:22:56<6:10:29,  1.8
+0: {'loss': 0.8649, 'grad_norm': 0.8253635577617087, 'learning_rate': 2e-05, 'epoch': 0.18}
+0: 4s/it]
 18%|█▊        | 2637/14725 [1:22:58<6:07:14,  1.82s/it]
 18%|█▊        | 2638/14725 [1:23:00<6:10:03,  1.84s/it]
 18%|█▊        | 2639/14725 [1:23:02<6:08:26,  1.83s/it]
 18%|█▊        | 2640/14725 [1:23:04<6:08:48,  1.83s/it]
                                                        

 18%|█▊        | 2640/14725 [1:23:04<6:08:48,  1.83s/it]
 18%|█▊        | 2641/14725 [1:23:06<6:14:12,  1.86s/it]
 18%|█▊        | 2642/14725 [1:23:07<6:11:24,  1.84s/it]
 18%|█▊        | 2643/14725 [1:23:09<6:08:06,  1.83s/it]
 18%|█▊        | 2644/14725 [1:23:11<6:03:37,  1.81s/it]
 18%|█▊        | 2645/14725 [1:23:13<6:05:06,  1.81s/it]
 18%|█▊        | 2646/14725 [1:23:14<6:02:16,  1.80s/it]
 18%|█▊        | 2647/14725 [1:23:16<6:03:30,  1.81s/it]
 18%|█▊        | 2648/14725 [1:23:18<6:08:19,  1.83s/it]
 18%|█▊        | 2649/14725 [1:23:20<6:05:22,  1.82s/it]
 18%|█▊        | 2650/14725 [1:23:22<6:04:41,  1.81s/it]
                                            
+0: {'loss': 0.8711, 'grad_norm': 0.7588096195478756, 'learning_rate': 2e-05, 'epoch': 0.18}
+0: {'loss': 0.8583, 'grad_norm': 0.8415845475149367, 'learning_rate': 2e-05, 'epoch': 0.18}
+0:             

 18%|█▊        | 2650/14725 [1:23:22<6:04:41,  1.81s/it]
 18%|█▊        | 2651/14725 [1:23:24<6:02:01,  1.80s/it]
 18%|█▊        | 2652/14725 [1:23:25<6:01:57,  1.80s/it]
 18%|█▊        | 2653/14725 [1:23:27<6:04:35,  1.81s/it]
 18%|█▊        | 2654/14725 [1:23:29<6:07:02,  1.82s/it]
 18%|█▊        | 2655/14725 [1:23:31<6:06:02,  1.82s/it]
 18%|█▊        | 2656/14725 [1:23:33<6:03:18,  1.81s/it]
 18%|█▊        | 2657/14725 [1:23:35<6:09:23,  1.84s/it]
 18%|█▊        | 2658/14725 [1:23:36<6:07:49,  1.83s/it]
 18%|█▊        | 2659/14725 [1:23:38<6:06:04,  1.82s/it]
 18%|█▊        | 2660/14725 [1:23:40<6:10:14,  1.84s/it]
                                                        

 18%|█▊        | 2660/14725 [1:23:40<6:10:14,  1.84s/it]
 18%|█▊        | 2661/14725 [1:23:42<6:11:42,  1.85s/it]
 18%|█▊        | 2662/14725 [1:23:44<6:08:25,  1.83s/it]
 18%|█▊        | 2663/14725 [1:23:46<6:11:08,  1.85s/it]
 18%|█▊        | 2664/14725 [1:23
+0: {'loss': 0.8817, 'grad_norm': 0.9005093294708203, 'learning_rate': 2e-05, 'epoch': 0.18}
+0: :47<6:13:24,  1.86s/it]
 18%|█▊        | 2665/14725 [1:23:49<6:07:54,  1.83s/it]
 18%|█▊        | 2666/14725 [1:23:51<6:09:37,  1.84s/it]
 18%|█▊        | 2667/14725 [1:23:53<6:08:33,  1.83s/it]
 18%|█▊        | 2668/14725 [1:23:55<6:08:50,  1.84s/it]
 18%|█▊        | 2669/14725 [1:23:56<6:05:07,  1.82s/it]
 18%|█▊        | 2670/14725 [1:23:58<6:04:01,  1.81s/it]
                                                        

 18%|█▊        | 2670/14725 [1:23:58<6:04:01,  1.81s/it]
 18%|█▊        | 2671/14725 [1:24:00<6:04:22,  1.81s/it]
 18%|█▊        | 2672/14725 [1:24:02<6:04:54,  1.82s/it]
 18%|█▊        | 2673/14725 [1:24:04<6:09:54,  1.84s/it]
 18%|█▊        | 2674/14725 [1:24:06<6:04:10,  1.81s/it]
 18%|█▊        | 2675/14725 [1:24:07<6:07:24,  1.83s/it]
 18%|█▊        | 2676/14725 [1:24:09<6:03:03,  1.81s/it]
 18%|█▊        | 2677/14725 [1:24:11<6:00:37,  1.80s/it]
 18%|█▊        | 2678/14725 [1:24:13<6:03:38,  1.81s/it]
 18%|█▊        | 2679/1
+0: {'loss': 0.8817, 'grad_norm': 0.7924035861050907, 'learning_rate': 2e-05, 'epoch': 0.18}
+0: {'loss': 0.8642, 'grad_norm': 0.7957729257883205, 'learning_rate': 2e-05, 'epoch': 0.18}
+0: 4725 [1:24:15<6:06:25,  1.83s/it]
 18%|█▊        | 2680/14725 [1:24:16<6:06:02,  1.82s/it]
                                                        

 18%|█▊        | 2680/14725 [1:24:16<6:06:02,  1.82s/it]
 18%|█▊        | 2681/14725 [1:24:18<6:04:04,  1.81s/it]
 18%|█▊        | 2682/14725 [1:24:20<6:00:19,  1.80s/it]
 18%|█▊        | 2683/14725 [1:24:22<6:00:56,  1.80s/it]
 18%|█▊        | 2684/14725 [1:24:24<6:03:09,  1.81s/it]
 18%|█▊        | 2685/14725 [1:24:25<6:01:11,  1.80s/it]
 18%|█▊        | 2686/14725 [1:24:27<5:59:21,  1.79s/it]
 18%|█▊        | 2687/14725 [1:24:29<6:01:05,  1.80s/it]
 18%|█▊        | 2688/14725 [1:24:31<6:01:20,  1.80s/it]
 18%|█▊        | 2689/14725 [1:24:33<5:59:10,  1.79s/it]
 18%|█▊        | 2690/14725 [1:24:34<5:57:03,  1.78s/it]
                                                        

 18%|█▊        | 2690/14725 [1:24:34<5:57:03,  1.78s/it]
 18%|█▊        | 2691/14725 [1:24:36<5:56:54,  1.78s/it]
 18%|█▊        |
+0: {'loss': 0.8648, 'grad_norm': 0.7663319565425148, 'learning_rate': 2e-05, 'epoch': 0.18}
+0:  2692/14725 [1:24:38<5:57:23,  1.78s/it]
 18%|█▊        | 2693/14725 [1:24:40<5:55:55,  1.77s/it]
 18%|█▊        | 2694/14725 [1:24:42<6:02:03,  1.81s/it]
 18%|█▊        | 2695/14725 [1:24:43<6:03:06,  1.81s/it]
 18%|█▊        | 2696/14725 [1:24:45<6:04:49,  1.82s/it]
 18%|█▊        | 2697/14725 [1:24:47<6:07:47,  1.83s/it]
 18%|█▊        | 2698/14725 [1:24:49<6:04:31,  1.82s/it]
 18%|█▊        | 2699/14725 [1:24:51<6:05:09,  1.82s/it]
 18%|█▊        | 2700/14725 [1:24:52<6:02:29,  1.81s/it]
                                                        

 18%|█▊        | 2700/14725 [1:24:53<6:02:29,  1.81s/it]
 18%|█▊        | 2701/14725 [1:24:54<6:06:40,  1.83s/it]
 18%|█▊        | 2702/14725 [1:24:56<6:06:46,  1.83s/it]
 18%|█▊        | 2703/14725 [1:24:58<6:05:06,  1.82s/it]
 18%|█▊        | 2704/14725 [1:25:00<6:04:03,  1.82s/it]
 18%|█▊        | 2705/14725 [1:25:02<6:00:53,  1.80s/it]
 18%|█▊        | 2706/14725 [1:25:03<6:02:07,  1.81s/it]
 18%|█�
+0: {'loss': 0.8673, 'grad_norm': 0.7831265312657889, 'learning_rate': 2e-05, 'epoch': 0.18}
+0: {'loss': 0.8746, 'grad_norm': 0.8586339985583546, 'learning_rate': 2e-05, 'epoch': 0.18}
+0: �        | 2707/14725 [1:25:05<6:03:19,  1.81s/it]
 18%|█▊        | 2708/14725 [1:25:07<6:00:23,  1.80s/it]
 18%|█▊        | 2709/14725 [1:25:09<5:58:37,  1.79s/it]
 18%|█▊        | 2710/14725 [1:25:11<5:59:03,  1.79s/it]
                                                        

 18%|█▊        | 2710/14725 [1:25:11<5:59:03,  1.79s/it]
 18%|█▊        | 2711/14725 [1:25:12<5:58:42,  1.79s/it]
 18%|█▊        | 2712/14725 [1:25:14<5:59:48,  1.80s/it]
 18%|█▊        | 2713/14725 [1:25:16<6:03:43,  1.82s/it]
 18%|█▊        | 2714/14725 [1:25:18<5:59:58,  1.80s/it]
 18%|█▊        | 2715/14725 [1:25:20<6:04:48,  1.82s/it]
 18%|█▊        | 2716/14725 [1:25:21<6:01:24,  1.81s/it]
 18%|█▊        | 2717/14725 [1:25:23<6:07:40,  1.84s/it]
 18%|█▊        | 2718/14725 [1:25:25<6:04:31,  1.82s/it]
 18%|█▊        | 2719/14725 [1:25:27<6:06:31,  1.83s/it]
 18%|█▊        | 2720/14725 [1:25:29<6:03:30,  1.82s/it]
                                                        

 18
+0: {'loss': 0.8595, 'grad_norm': 0.7896434631826281, 'learning_rate': 2e-05, 'epoch': 0.19}
+0: %|█▊        | 2720/14725 [1:25:29<6:03:30,  1.82s/it]
 18%|█▊        | 2721/14725 [1:25:31<6:03:36,  1.82s/it]
 18%|█▊        | 2722/14725 [1:25:32<6:05:30,  1.83s/it]
 18%|█▊        | 2723/14725 [1:25:34<6:07:41,  1.84s/it]
 18%|█▊        | 2724/14725 [1:25:36<6:03:36,  1.82s/it]
 19%|█▊        | 2725/14725 [1:25:38<6:00:11,  1.80s/it]
 19%|█▊        | 2726/14725 [1:25:40<5:59:19,  1.80s/it]
 19%|█▊        | 2727/14725 [1:25:41<6:02:31,  1.81s/it]
 19%|█▊        | 2728/14725 [1:25:43<6:01:20,  1.81s/it]
 19%|█▊        | 2729/14725 [1:25:45<6:01:00,  1.81s/it]
 19%|█▊        | 2730/14725 [1:25:47<6:01:19,  1.81s/it]
                                                        

 19%|█▊        | 2730/14725 [1:25:47<6:01:19,  1.81s/it]
 19%|█▊        | 2731/14725 [1:25:49<6:00:52,  1.81s/it]
 19%|█▊        | 2732/14725 [1:25:50<5:57:58,  1.79s/it]
 19%|█▊        | 2733/14725 [1:25:52<6:00:34,  1.80s/it]
 19%|█▊        | 2734/14725 [1:25:54<6:07:18,  1.8
+0: {'loss': 0.8741, 'grad_norm': 0.7786336893382161, 'learning_rate': 2e-05, 'epoch': 0.19}
+0: 4s/it]
 19%|█▊        | 2735/14725 [1:25:56<6:12:27,  1.86s/it]
 19%|█▊        | 2736/14725 [1:25:58<6:09:34,  1.85s/it]
 19%|█▊        | 2737/14725 [1:26:00<6:10:57,  1.86s/it]
 19%|█▊        | 2738/14725 [1:26:02<6:10:06,  1.85s/it]
 19%|█▊        | 2739/14725 [1:26:03<6:05:48,  1.83s/it]
 19%|█▊        | 2740/14725 [1:26:05<6:05:51,  1.83s/it]
                                                        

 19%|█▊        | 2740/14725 [1:26:05<6:05:51,  1.83s/it]
 19%|█▊        | 2741/14725 [1:26:07<6:10:09,  1.85s/it]
 19%|█▊        | 2742/14725 [1:26:09<6:13:23,  1.87s/it]
 19%|█▊        | 2743/14725 [1:26:11<6:08:02,  1.84s/it]
 19%|█▊        | 2744/14725 [1:26:13<6:06:50,  1.84s/it]
 19%|█▊        | 2745/14725 [1:26:15<6:06:47,  1.84s/it]
 19%|█▊        | 2746/14725 [1:26:16<6:03:13,  1.82s/it]
 19%|█▊        | 2747/14725 [1:26:18<6:02:30,  1.82s/it]
 19%|█▊        | 2748/14725 [1:26:20<6:07:11,  1.84s/it]
 19%|█▊        | 2749/14725 [1:26:22<6:0
+0: {'loss': 0.8656, 'grad_norm': 0.8946492015543974, 'learning_rate': 2e-05, 'epoch': 0.19}
+0: {'loss': 0.8596, 'grad_norm': 0.7921905573607437, 'learning_rate': 2e-05, 'epoch': 0.19}
+0: 6:18,  1.84s/it]
 19%|█▊        | 2750/14725 [1:26:24<6:02:39,  1.82s/it]
                                                        

 19%|█▊        | 2750/14725 [1:26:24<6:02:39,  1.82s/it]
 19%|█▊        | 2751/14725 [1:26:26<6:09:49,  1.85s/it]
 19%|█▊        | 2752/14725 [1:26:27<6:08:11,  1.85s/it]
 19%|█▊        | 2753/14725 [1:26:29<6:06:29,  1.84s/it]
 19%|█▊        | 2754/14725 [1:26:31<6:02:27,  1.82s/it]
 19%|█▊        | 2755/14725 [1:26:33<5:59:24,  1.80s/it]
 19%|█▊        | 2756/14725 [1:26:34<5:59:19,  1.80s/it]
 19%|█▊        | 2757/14725 [1:26:36<5:59:42,  1.80s/it]
 19%|█▊        | 2758/14725 [1:26:38<6:00:02,  1.81s/it]
 19%|█▊        | 2759/14725 [1:26:40<6:00:03,  1.81s/it]
 19%|█▊        | 2760/14725 [1:26:42<5:59:06,  1.80s/it]
                                                        

 19%|█▊        | 2760/14725 [1:26:42<5:59:06,  1.80s/it]
 19%|█▉        | 2761/14725 [1:26:44<6:01:42,  1.81s/it]
 19%|█▉        | 2762/14725 [1:26
+0: {'loss': 0.8543, 'grad_norm': 0.7787400843194494, 'learning_rate': 2e-05, 'epoch': 0.19}
+0: :45<6:06:25,  1.84s/it]
 19%|█▉        | 2763/14725 [1:26:47<6:07:35,  1.84s/it]
 19%|█▉        | 2764/14725 [1:26:49<6:07:19,  1.84s/it]
 19%|█▉        | 2765/14725 [1:26:51<6:02:34,  1.82s/it]
 19%|█▉        | 2766/14725 [1:26:53<6:07:16,  1.84s/it]
 19%|█▉        | 2767/14725 [1:26:55<6:02:17,  1.82s/it]
 19%|█▉        | 2768/14725 [1:26:56<5:59:35,  1.80s/it]
 19%|█▉        | 2769/14725 [1:26:58<6:03:03,  1.82s/it]
 19%|█▉        | 2770/14725 [1:27:00<6:04:20,  1.83s/it]
                                                        

 19%|█▉        | 2770/14725 [1:27:00<6:04:20,  1.83s/it]
 19%|█▉        | 2771/14725 [1:27:02<6:06:13,  1.84s/it]
 19%|█▉        | 2772/14725 [1:27:04<6:05:38,  1.84s/it]
 19%|█▉        | 2773/14725 [1:27:06<6:02:54,  1.82s/it]
 19%|█▉        | 2774/14725 [1:27:07<5:59:08,  1.80s/it]
 19%|█▉        | 2775/14725 [1:27:09<6:06:01,  1.84s/it]
 19%|█▉        | 2776/14725 [1:27:11<6:08:18,  1.85s/it]
 19%|█▉        | 2777/1
+0: {'loss': 0.879, 'grad_norm': 0.8370129593849817, 'learning_rate': 2e-05, 'epoch': 0.19}
+0: {'loss': 0.8666, 'grad_norm': 0.8163100790110679, 'learning_rate': 2e-05, 'epoch': 0.19}
+0: 4725 [1:27:13<6:06:56,  1.84s/it]
 19%|█▉        | 2778/14725 [1:27:15<6:06:39,  1.84s/it]
 19%|█▉        | 2779/14725 [1:27:17<6:02:14,  1.82s/it]
 19%|█▉        | 2780/14725 [1:27:18<6:04:25,  1.83s/it]
                                                        

 19%|█▉        | 2780/14725 [1:27:18<6:04:25,  1.83s/it]
 19%|█▉        | 2781/14725 [1:27:20<6:02:21,  1.82s/it]
 19%|█▉        | 2782/14725 [1:27:22<5:59:14,  1.80s/it]
 19%|█▉        | 2783/14725 [1:27:24<6:00:09,  1.81s/it]
 19%|█▉        | 2784/14725 [1:27:26<6:02:07,  1.82s/it]
 19%|█▉        | 2785/14725 [1:27:27<6:02:01,  1.82s/it]
 19%|█▉        | 2786/14725 [1:27:29<6:04:58,  1.83s/it]
 19%|█▉        | 2787/14725 [1:27:31<6:08:18,  1.85s/it]
 19%|█▉        | 2788/14725 [1:27:33<6:08:06,  1.85s/it]
 19%|█▉        | 2789/14725 [1:27:35<6:02:26,  1.82s/it]
 19%|█▉        | 2790/14725 [1:27:37<6:06:39,  1.84s/it]
                                                        

 19%|█▉        |
+0: {'loss': 0.8807, 'grad_norm': 0.8266610524827384, 'learning_rate': 2e-05, 'epoch': 0.19}
+0:  2790/14725 [1:27:37<6:06:39,  1.84s/it]
 19%|█▉        | 2791/14725 [1:27:38<6:01:08,  1.82s/it]
 19%|█▉        | 2792/14725 [1:27:40<6:04:37,  1.83s/it]
 19%|█▉        | 2793/14725 [1:27:42<6:01:50,  1.82s/it]
 19%|█▉        | 2794/14725 [1:27:44<5:59:27,  1.81s/it]
 19%|█▉        | 2795/14725 [1:27:46<5:59:57,  1.81s/it]
 19%|█▉        | 2796/14725 [1:27:47<5:56:40,  1.79s/it]
 19%|█▉        | 2797/14725 [1:27:49<5:54:17,  1.78s/it]
 19%|█▉        | 2798/14725 [1:27:51<5:53:22,  1.78s/it]
 19%|█▉        | 2799/14725 [1:27:53<5:52:02,  1.77s/it]
 19%|█▉        | 2800/14725 [1:27:55<5:52:53,  1.78s/it]
                                                        

 19%|█▉        | 2800/14725 [1:27:55<5:52:53,  1.78s/it]
 19%|█▉        | 2801/14725 [1:27:56<5:53:45,  1.78s/it]
 19%|█▉        | 2802/14725 [1:27:58<5:57:12,  1.80s/it]
 19%|█▉        | 2803/14725 [1:28:00<5:56:04,  1.79s/it]
 19%|█▉        | 2804/14725 [1:28:02<5:55:47,  1.79s/it]
 19%|█�
+0: {'loss': 0.8601, 'grad_norm': 0.8898460566299577, 'learning_rate': 2e-05, 'epoch': 0.19}
+0: �        | 2805/14725 [1:28:03<5:54:49,  1.79s/it]
 19%|█▉        | 2806/14725 [1:28:05<5:59:22,  1.81s/it]
 19%|█▉        | 2807/14725 [1:28:07<6:02:22,  1.82s/it]
 19%|█▉        | 2808/14725 [1:28:09<5:57:54,  1.80s/it]
 19%|█▉        | 2809/14725 [1:28:11<5:56:59,  1.80s/it]
 19%|█▉        | 2810/14725 [1:28:12<5:54:04,  1.78s/it]
                                                        

 19%|█▉        | 2810/14725 [1:28:12<5:54:04,  1.78s/it]
 19%|█▉        | 2811/14725 [1:28:14<6:01:45,  1.82s/it]
 19%|█▉        | 2812/14725 [1:28:16<6:04:05,  1.83s/it]
 19%|█▉        | 2813/14725 [1:28:18<6:05:08,  1.84s/it]
 19%|█▉        | 2814/14725 [1:28:20<6:11:42,  1.87s/it]
 19%|█▉        | 2815/14725 [1:28:22<6:06:33,  1.85s/it]
 19%|█▉        | 2816/14725 [1:28:24<6:02:45,  1.83s/it]
 19%|█▉        | 2817/14725 [1:28:25<5:59:09,  1.81s/it]
 19%|█▉        | 2818/14725 [1:28:27<5:57:33,  1.80s/it]
 19%|█▉        | 2819/14725 [1:28:29<5:59:12,  1.81s/it]
+0: {'loss': 0.8494, 'grad_norm': 0.8355137062647691, 'learning_rate': 2e-05, 'epoch': 0.19}
+0: {'loss': 0.8592, 'grad_norm': 0.7485890053503196, 'learning_rate': 2e-05, 'epoch': 0.19}
+0:  19%|█▉        | 2820/14725 [1:28:31<5:59:15,  1.81s/it]
                                                        

 19%|█▉        | 2820/14725 [1:28:31<5:59:15,  1.81s/it]
 19%|█▉        | 2821/14725 [1:28:33<5:59:40,  1.81s/it]
 19%|█▉        | 2822/14725 [1:28:35<6:03:07,  1.83s/it]
 19%|█▉        | 2823/14725 [1:28:36<6:03:18,  1.83s/it]
 19%|█▉        | 2824/14725 [1:28:38<6:00:18,  1.82s/it]
 19%|█▉        | 2825/14725 [1:28:40<5:57:44,  1.80s/it]
 19%|█▉        | 2826/14725 [1:28:42<6:00:31,  1.82s/it]
 19%|█▉        | 2827/14725 [1:28:44<5:58:17,  1.81s/it]
 19%|█▉        | 2828/14725 [1:28:45<5:59:49,  1.81s/it]
 19%|█▉        | 2829/14725 [1:28:47<5:57:09,  1.80s/it]
 19%|█▉        | 2830/14725 [1:28:49<5:55:50,  1.79s/it]
                                                        

 19%|█▉        | 2830/14725 [1:28:49<5:55:50,  1.79s/it]
 19%|█▉        | 2831/14725 [1:28:51<6:00:45,  1.82s/it]
 19%|█▉        | 2832/14725 [1:28:53<6:03:07,  1.8
+0: {'loss': 0.8553, 'grad_norm': 0.7832195844590819, 'learning_rate': 2e-05, 'epoch': 0.19}
+0: 3s/it]
 19%|█▉        | 2833/14725 [1:28:54<5:59:29,  1.81s/it]
 19%|█▉        | 2834/14725 [1:28:56<6:04:54,  1.84s/it]
 19%|█▉        | 2835/14725 [1:28:58<6:02:45,  1.83s/it]
 19%|█▉        | 2836/14725 [1:29:00<5:59:35,  1.81s/it]
 19%|█▉        | 2837/14725 [1:29:02<5:58:12,  1.81s/it]
 19%|█▉        | 2838/14725 [1:29:04<5:57:58,  1.81s/it]
 19%|█▉        | 2839/14725 [1:29:05<5:56:10,  1.80s/it]
 19%|█▉        | 2840/14725 [1:29:07<6:00:11,  1.82s/it]
                                                        

 19%|█▉        | 2840/14725 [1:29:07<6:00:11,  1.82s/it]
 19%|█▉        | 2841/14725 [1:29:09<6:05:55,  1.85s/it]
 19%|█▉        | 2842/14725 [1:29:11<6:04:33,  1.84s/it]
 19%|█▉        | 2843/14725 [1:29:13<6:02:24,  1.83s/it]
 19%|█▉        | 2844/14725 [1:29:15<6:00:27,  1.82s/it]
 19%|█▉        | 2845/14725 [1:29:16<6:01:15,  1.82s/it]
 19%|█▉        | 2846/14725 [1:29:18<6:07:50,  1.86s/it]
 19%|█▉        | 2847/14725 [1:29:20<6:0
+0: {'loss': 0.9004, 'grad_norm': 0.8675912739814656, 'learning_rate': 2e-05, 'epoch': 0.19}
+0: {'loss': 0.8852, 'grad_norm': 0.81068263539733, 'learning_rate': 2e-05, 'epoch': 0.19}
+0: 3:22,  1.84s/it]
 19%|█▉        | 2848/14725 [1:29:22<6:00:34,  1.82s/it]
 19%|█▉        | 2849/14725 [1:29:24<6:03:10,  1.83s/it]
 19%|█▉        | 2850/14725 [1:29:25<6:00:30,  1.82s/it]
                                                        

 19%|█▉        | 2850/14725 [1:29:26<6:00:30,  1.82s/it]
 19%|█▉        | 2851/14725 [1:29:27<6:00:58,  1.82s/it]
 19%|█▉        | 2852/14725 [1:29:29<6:01:03,  1.82s/it]
 19%|█▉        | 2853/14725 [1:29:31<5:57:21,  1.81s/it]
 19%|█▉        | 2854/14725 [1:29:33<5:55:23,  1.80s/it]
 19%|█▉        | 2855/14725 [1:29:35<5:56:37,  1.80s/it]
 19%|█▉        | 2856/14725 [1:29:36<5:55:04,  1.79s/it]
 19%|█▉        | 2857/14725 [1:29:38<5:54:10,  1.79s/it]
 19%|█▉        | 2858/14725 [1:29:40<5:55:09,  1.80s/it]
 19%|█▉        | 2859/14725 [1:29:42<5:54:57,  1.79s/it]
 19%|█▉        | 2860/14725 [1:29:43<5:53:24,  1.79s/it]
                                                        

 19%|█▉        | 2860/14725 [1:29
+0: {'loss': 0.8833, 'grad_norm': 0.7605288817087981, 'learning_rate': 2e-05, 'epoch': 0.19}
+0: :43<5:53:24,  1.79s/it]
 19%|█▉        | 2861/14725 [1:29:45<5:54:14,  1.79s/it]
 19%|█▉        | 2862/14725 [1:29:47<5:56:53,  1.81s/it]
 19%|█▉        | 2863/14725 [1:29:49<5:54:39,  1.79s/it]
 19%|█▉        | 2864/14725 [1:29:51<5:55:47,  1.80s/it]
 19%|█▉        | 2865/14725 [1:29:53<6:01:21,  1.83s/it]
 19%|█▉        | 2866/14725 [1:29:54<5:58:44,  1.82s/it]
 19%|█▉        | 2867/14725 [1:29:56<6:05:53,  1.85s/it]
 19%|█▉        | 2868/14725 [1:29:58<6:05:29,  1.85s/it]
 19%|█▉        | 2869/14725 [1:30:00<6:05:05,  1.85s/it]
 19%|█▉        | 2870/14725 [1:30:02<6:01:43,  1.83s/it]
                                                        

 19%|█▉        | 2870/14725 [1:30:02<6:01:43,  1.83s/it]
 19%|█▉        | 2871/14725 [1:30:04<5:58:02,  1.81s/it]
 20%|█▉        | 2872/14725 [1:30:05<5:59:42,  1.82s/it]
 20%|█▉        | 2873/14725 [1:30:07<6:00:35,  1.83s/it]
 20%|█▉        | 2874/14725 [1:30:09<6:02:07,  1.83s/it]
 20%|█▉        | 2875/1
+0: {'loss': 0.8691, 'grad_norm': 0.7823850890394428, 'learning_rate': 2e-05, 'epoch': 0.2}
+0: 4725 [1:30:11<6:02:25,  1.84s/it]
 20%|█▉        | 2876/14725 [1:30:13<6:02:25,  1.84s/it]
 20%|█▉        | 2877/14725 [1:30:15<5:59:14,  1.82s/it]
 20%|█▉        | 2878/14725 [1:30:16<5:55:34,  1.80s/it]
 20%|█▉        | 2879/14725 [1:30:18<5:58:56,  1.82s/it]
 20%|█▉        | 2880/14725 [1:30:20<5:57:41,  1.81s/it]
                                                        

 20%|█▉        | 2880/14725 [1:30:20<5:57:41,  1.81s/it]
 20%|█▉        | 2881/14725 [1:30:22<5:56:39,  1.81s/it]
 20%|█▉        | 2882/14725 [1:30:23<5:55:20,  1.80s/it]
 20%|█▉        | 2883/14725 [1:30:25<5:54:52,  1.80s/it]
 20%|█▉        | 2884/14725 [1:30:27<5:53:59,  1.79s/it]
 20%|█▉        | 2885/14725 [1:30:29<5:54:10,  1.79s/it]
 20%|█▉        | 2886/14725 [1:30:31<5:53:03,  1.79s/it]
 20%|█▉        | 2887/14725 [1:30:33<5:57:22,  1.81s/it]
 20%|█▉        | 2888/14725 [1:30:34<5:56:10,  1.81s/it]
 20%|█▉        | 2889/14725 [1:30:36<5:53:59,  1.79s/it]
 20%|█▉      
+0: {'loss': 0.87, 'grad_norm': 0.8185249103887866, 'learning_rate': 2e-05, 'epoch': 0.2}
+0: {'loss': 0.8815, 'grad_norm': 0.78428658136021, 'learning_rate': 2e-05, 'epoch': 0.2}
+0:   | 2890/14725 [1:30:38<5:53:09,  1.79s/it]
                                                        

 20%|█▉        | 2890/14725 [1:30:38<5:53:09,  1.79s/it]
 20%|█▉        | 2891/14725 [1:30:40<5:56:25,  1.81s/it]
 20%|█▉        | 2892/14725 [1:30:42<6:01:09,  1.83s/it]
 20%|█▉        | 2893/14725 [1:30:43<6:05:54,  1.86s/it]
 20%|█▉        | 2894/14725 [1:30:45<6:07:39,  1.86s/it]
 20%|█▉        | 2895/14725 [1:30:47<6:04:50,  1.85s/it]
 20%|█▉        | 2896/14725 [1:30:49<6:02:38,  1.84s/it]
 20%|█▉        | 2897/14725 [1:30:51<6:00:04,  1.83s/it]
 20%|█▉        | 2898/14725 [1:30:53<6:00:46,  1.83s/it]
 20%|█▉        | 2899/14725 [1:30:55<6:04:48,  1.85s/it]
 20%|█▉        | 2900/14725 [1:30:56<6:00:21,  1.83s/it]
                                                        

 20%|█▉        | 2900/14725 [1:30:56<6:00:21,  1.83s/it]
 20%|█▉        | 2901/14725 [1:30:58<6:01:32,  1.83s/it]
 20%|█▉        | 2902/14725 [1:31:00<6:01:18,  1.83s/it]
 20%|█�
+0: {'loss': 0.8768, 'grad_norm': 0.7823462187746801, 'learning_rate': 2e-05, 'epoch': 0.2}
+0: �        | 2903/14725 [1:31:02<5:57:05,  1.81s/it]
 20%|█▉        | 2904/14725 [1:31:04<5:59:48,  1.83s/it]
 20%|█▉        | 2905/14725 [1:31:05<5:55:37,  1.81s/it]
 20%|█▉        | 2906/14725 [1:31:07<5:59:39,  1.83s/it]
 20%|█▉        | 2907/14725 [1:31:09<5:56:53,  1.81s/it]
 20%|█▉        | 2908/14725 [1:31:11<5:54:05,  1.80s/it]
 20%|█▉        | 2909/14725 [1:31:13<5:52:01,  1.79s/it]
 20%|█▉        | 2910/14725 [1:31:15<6:01:06,  1.83s/it]
                                                        

 20%|█▉        | 2910/14725 [1:31:15<6:01:06,  1.83s/it]
 20%|█▉        | 2911/14725 [1:31:16<5:56:29,  1.81s/it]
 20%|█▉        | 2912/14725 [1:31:18<5:56:22,  1.81s/it]
 20%|█▉        | 2913/14725 [1:31:20<5:58:28,  1.82s/it]
 20%|█▉        | 2914/14725 [1:31:22<5:56:14,  1.81s/it]
 20%|█▉        | 2915/14725 [1:31:23<5:54:56,  1.80s/it]
 20%|█▉        | 2916/14725 [1:31:25<5:51:44,  1.79s/it]
 20%|█▉        | 2917/14725 [1:31:27<5:55:00,  1.80s/it]
+0: {'loss': 0.8689, 'grad_norm': 0.8477430945387907, 'learning_rate': 2e-05, 'epoch': 0.2}
+0: {'loss': 0.8941, 'grad_norm': 0.7674030412094545, 'learning_rate': 2e-05, 'epoch': 0.2}
+0:  20%|█▉        | 2918/14725 [1:31:29<5:55:13,  1.81s/it]
 20%|█▉        | 2919/14725 [1:31:31<5:56:01,  1.81s/it]
 20%|█▉        | 2920/14725 [1:31:33<5:55:28,  1.81s/it]
                                                        

 20%|█▉        | 2920/14725 [1:31:33<5:55:28,  1.81s/it]
 20%|█▉        | 2921/14725 [1:31:34<6:00:24,  1.83s/it]
 20%|█▉        | 2922/14725 [1:31:36<6:04:12,  1.85s/it]
 20%|█▉        | 2923/14725 [1:31:38<5:58:56,  1.82s/it]
 20%|█▉        | 2924/14725 [1:31:40<5:57:12,  1.82s/it]
 20%|█▉        | 2925/14725 [1:31:42<5:54:23,  1.80s/it]
 20%|█▉        | 2926/14725 [1:31:43<5:51:20,  1.79s/it]
 20%|█▉        | 2927/14725 [1:31:45<5:51:57,  1.79s/it]
 20%|█▉        | 2928/14725 [1:31:47<5:51:57,  1.79s/it]
 20%|█▉        | 2929/14725 [1:31:49<5:52:55,  1.80s/it]
 20%|█▉        | 2930/14725 [1:31:51<5:57:46,  1.82s/it]
                                                        

 20%|█▉        | 2930/14725 [1:31:51<5:57:46,  1.8
+0: {'loss': 0.8559, 'grad_norm': 0.80077529724608, 'learning_rate': 2e-05, 'epoch': 0.2}
+0: 2s/it]
 20%|█▉        | 2931/14725 [1:31:53<6:04:12,  1.85s/it]
 20%|█▉        | 2932/14725 [1:31:54<6:02:30,  1.84s/it]
 20%|█▉        | 2933/14725 [1:31:56<5:58:01,  1.82s/it]
 20%|█▉        | 2934/14725 [1:31:58<5:59:22,  1.83s/it]
 20%|█▉        | 2935/14725 [1:32:00<5:56:43,  1.82s/it]
 20%|█▉        | 2936/14725 [1:32:02<5:54:47,  1.81s/it]
 20%|█▉        | 2937/14725 [1:32:03<5:54:27,  1.80s/it]
 20%|█▉        | 2938/14725 [1:32:05<5:56:02,  1.81s/it]
 20%|█▉        | 2939/14725 [1:32:07<5:56:24,  1.81s/it]
 20%|█▉        | 2940/14725 [1:32:09<5:54:24,  1.80s/it]
                                                        

 20%|█▉        | 2940/14725 [1:32:09<5:54:24,  1.80s/it]
 20%|█▉        | 2941/14725 [1:32:11<5:56:00,  1.81s/it]
 20%|█▉        | 2942/14725 [1:32:12<5:54:31,  1.81s/it]
 20%|█▉        | 2943/14725 [1:32:14<5:52:46,  1.80s/it]
 20%|█▉        | 2944/14725 [1:32:16<5:53:26,  1.80s/it]
 20%|██        | 2945/14725 [1:32:18<6:0
+0: {'loss': 0.8395, 'grad_norm': 0.7958410131025193, 'learning_rate': 2e-05, 'epoch': 0.2}
+0: 1:54,  1.84s/it]
 20%|██        | 2946/14725 [1:32:23<9:15:21,  2.83s/it]
 20%|██        | 2947/14725 [1:32:25<8:14:22,  2.52s/it]
 20%|██        | 2948/14725 [1:32:27<7:30:26,  2.29s/it]
 20%|██        | 2949/14725 [1:32:28<7:00:06,  2.14s/it]
 20%|██        | 2950/14725 [1:32:30<6:39:24,  2.04s/it]
                                                        

 20%|██        | 2950/14725 [1:32:30<6:39:24,  2.04s/it]
 20%|██        | 2951/14725 [1:32:32<6:28:44,  1.98s/it]
 20%|██        | 2952/14725 [1:32:34<6:17:44,  1.93s/it]
 20%|██        | 2953/14725 [1:32:36<6:07:48,  1.87s/it]
 20%|██        | 2954/14725 [1:32:37<6:02:59,  1.85s/it]
 20%|██        | 2955/14725 [1:32:39<5:59:01,  1.83s/it]
 20%|██        | 2956/14725 [1:32:41<5:58:08,  1.83s/it]
 20%|██        | 2957/14725 [1:32:43<5:54:53,  1.81s/it]
 20%|██        | 2958/14725 [1:32:45<5:54:45,  1.81s/it]
 20%|██        | 2959/14725 [1:32:46<5:53:52,  1.80s/it]
 20%|██        | 2960/14725 [1
+0: {'loss': 0.8725, 'grad_norm': 0.934990141109737, 'learning_rate': 2e-05, 'epoch': 0.2}
+0: {'loss': 0.8411, 'grad_norm': 0.754823219444664, 'learning_rate': 2e-05, 'epoch': 0.2}
+0: :32:48<5:50:56,  1.79s/it]
                                                        

 20%|██        | 2960/14725 [1:32:48<5:50:56,  1.79s/it]
 20%|██        | 2961/14725 [1:32:50<5:54:37,  1.81s/it]
 20%|██        | 2962/14725 [1:32:52<5:52:23,  1.80s/it]
 20%|██        | 2963/14725 [1:32:54<5:54:37,  1.81s/it]
 20%|██        | 2964/14725 [1:32:55<5:56:22,  1.82s/it]
 20%|██        | 2965/14725 [1:32:57<5:55:55,  1.82s/it]
 20%|██        | 2966/14725 [1:32:59<5:54:53,  1.81s/it]
 20%|██        | 2967/14725 [1:33:01<5:57:18,  1.82s/it]
 20%|██        | 2968/14725 [1:33:03<5:59:38,  1.84s/it]
 20%|██        | 2969/14725 [1:33:05<5:55:34,  1.81s/it]
 20%|██        | 2970/14725 [1:33:06<5:51:59,  1.80s/it]
                                                        

 20%|██        | 2970/14725 [1:33:06<5:51:59,  1.80s/it]
 20%|██        | 2971/14725 [1:33:08<5:52:38,  1.80s/it]
 20%|██        | 2972/14725 [1:33:10<6:16:01,  1.92s/it]
 20%|██        | 2973/1
+0: {'loss': 0.8629, 'grad_norm': 0.804950351461703, 'learning_rate': 2e-05, 'epoch': 0.2}
+0: 4725 [1:33:12<6:14:22,  1.91s/it]
 20%|██        | 2974/14725 [1:33:14<6:08:45,  1.88s/it]
 20%|██        | 2975/14725 [1:33:16<6:07:09,  1.87s/it]
 20%|██        | 2976/14725 [1:33:18<6:03:04,  1.85s/it]
 20%|██        | 2977/14725 [1:33:19<6:00:37,  1.84s/it]
 20%|██        | 2978/14725 [1:33:21<5:57:03,  1.82s/it]
 20%|██        | 2979/14725 [1:33:23<5:59:27,  1.84s/it]
 20%|██        | 2980/14725 [1:33:25<5:56:17,  1.82s/it]
                                                        

 20%|██        | 2980/14725 [1:33:25<5:56:17,  1.82s/it]
 20%|██        | 2981/14725 [1:33:27<5:55:33,  1.82s/it]
 20%|██        | 2982/14725 [1:33:29<5:56:35,  1.82s/it]
 20%|██        | 2983/14725 [1:33:30<5:55:10,  1.81s/it]
 20%|██        | 2984/14725 [1:33:32<5:53:14,  1.81s/it]
 20%|██        | 2985/14725 [1:33:34<5:57:37,  1.83s/it]
 20%|██        | 2986/14725 [1:33:36<5:54:14,  1.81s/it]
 20%|██        | 2987/14725 [1:33:38<5:57:33,  1.83s/it]
 20%|██      
+0: {'loss': 0.8513, 'grad_norm': 0.7995836006847025, 'learning_rate': 2e-05, 'epoch': 0.2}
+0: {'loss': 0.8892, 'grad_norm': 0.8413696259646972, 'learning_rate': 2e-05, 'epoch': 0.2}
+0:   | 2988/14725 [1:33:40<5:59:42,  1.84s/it]
 20%|██        | 2989/14725 [1:33:41<6:00:00,  1.84s/it]
 20%|██        | 2990/14725 [1:33:43<5:55:24,  1.82s/it]
                                                        

 20%|██        | 2990/14725 [1:33:43<5:55:24,  1.82s/it]
 20%|██        | 2991/14725 [1:33:45<5:54:39,  1.81s/it]
 20%|██        | 2992/14725 [1:33:47<5:53:09,  1.81s/it]
 20%|██        | 2993/14725 [1:33:49<5:52:23,  1.80s/it]
 20%|██        | 2994/14725 [1:33:50<5:52:33,  1.80s/it]
 20%|██        | 2995/14725 [1:33:52<5:54:40,  1.81s/it]
 20%|██        | 2996/14725 [1:33:54<5:59:24,  1.84s/it]
 20%|██        | 2997/14725 [1:33:56<6:01:03,  1.85s/it]
 20%|██        | 2998/14725 [1:33:58<5:56:58,  1.83s/it]
 20%|██        | 2999/14725 [1:34:00<6:02:45,  1.86s/it]
 20%|██        | 3000/14725 [1:34:01<6:01:15,  1.85s/it]
                                                        

 20%|██        | 3000/14725 [1:34:01<6:01:15,  1.85s/it]
 20%|█�
+0: {'loss': 0.858, 'grad_norm': 0.8109283342106371, 'learning_rate': 2e-05, 'epoch': 0.2}
+0: �        | 3001/14725 [1:34:03<6:00:56,  1.85s/it]
 20%|██        | 3002/14725 [1:34:05<6:00:01,  1.84s/it]
 20%|██        | 3003/14725 [1:34:07<5:59:58,  1.84s/it]
 20%|██        | 3004/14725 [1:34:09<6:09:42,  1.89s/it]
 20%|██        | 3005/14725 [1:34:11<6:03:43,  1.86s/it]
 20%|██        | 3006/14725 [1:34:13<6:07:41,  1.88s/it]
 20%|██        | 3007/14725 [1:34:15<6:02:29,  1.86s/it]
 20%|██        | 3008/14725 [1:34:16<5:57:23,  1.83s/it]
 20%|██        | 3009/14725 [1:34:18<5:58:36,  1.84s/it]
 20%|██        | 3010/14725 [1:34:20<5:57:29,  1.83s/it]
                                                        

 20%|██        | 3010/14725 [1:34:20<5:57:29,  1.83s/it]
 20%|██        | 3011/14725 [1:34:22<5:55:12,  1.82s/it]
 20%|██        | 3012/14725 [1:34:24<5:54:04,  1.81s/it]
 20%|██        | 3013/14725 [1:34:25<5:51:26,  1.80s/it]
 20%|██        | 3014/14725 [1:34:27<5:51:21,  1.80s/it]
 20%|██        | 3015/14725 [1:34:29<5:50:52,  1.80s/it]
+0: {'loss': 0.8656, 'grad_norm': 0.8853050374351603, 'learning_rate': 2e-05, 'epoch': 0.21}
+0:  20%|██        | 3016/14725 [1:34:31<5:49:13,  1.79s/it]
 20%|██        | 3017/14725 [1:34:33<5:51:14,  1.80s/it]
 20%|██        | 3018/14725 [1:34:34<5:50:36,  1.80s/it]
 21%|██        | 3019/14725 [1:34:36<5:53:13,  1.81s/it]
 21%|██        | 3020/14725 [1:34:38<5:55:46,  1.82s/it]
                                                        

 21%|██        | 3020/14725 [1:34:38<5:55:46,  1.82s/it]
 21%|██        | 3021/14725 [1:34:40<5:54:40,  1.82s/it]
 21%|██        | 3022/14725 [1:34:42<5:54:20,  1.82s/it]
 21%|██        | 3023/14725 [1:34:43<5:51:58,  1.80s/it]
 21%|██        | 3024/14725 [1:34:45<5:54:05,  1.82s/it]
 21%|██        | 3025/14725 [1:34:47<5:51:24,  1.80s/it]
 21%|██        | 3026/14725 [1:34:49<5:52:24,  1.81s/it]
 21%|██        | 3027/14725 [1:34:51<5:52:48,  1.81s/it]
 21%|██        | 3028/14725 [1:34:52<5:51:12,  1.80s/it]
 21%|██        | 3029/14725 [1:34:54<5:49:57,  1.80s/it]
 21%|██        | 3030/14725 [1:34:56<5:49:14,  
+0: {'loss': 0.8698, 'grad_norm': 0.7980203287439689, 'learning_rate': 2e-05, 'epoch': 0.21}
+0: {'loss': 0.8669, 'grad_norm': 0.8856352368042745, 'learning_rate': 2e-05, 'epoch': 0.21}
+0: 1.79s/it]
                                                        

 21%|██        | 3030/14725 [1:34:56<5:49:14,  1.79s/it]
 21%|██        | 3031/14725 [1:34:58<5:49:49,  1.79s/it]
 21%|██        | 3032/14725 [1:35:00<5:47:35,  1.78s/it]
 21%|██        | 3033/14725 [1:35:01<5:47:07,  1.78s/it]
 21%|██        | 3034/14725 [1:35:03<5:44:57,  1.77s/it]
 21%|██        | 3035/14725 [1:35:05<5:50:01,  1.80s/it]
 21%|██        | 3036/14725 [1:35:07<5:48:15,  1.79s/it]
 21%|██        | 3037/14725 [1:35:09<5:51:05,  1.80s/it]
 21%|██        | 3038/14725 [1:35:10<5:48:42,  1.79s/it]
 21%|██        | 3039/14725 [1:35:12<5:48:54,  1.79s/it]
 21%|██        | 3040/14725 [1:35:14<5:47:34,  1.78s/it]
                                                        

 21%|██        | 3040/14725 [1:35:14<5:47:34,  1.78s/it]
 21%|██        | 3041/14725 [1:35:16<6:00:31,  1.85s/it]
 21%|██        | 3042/14725 [1:35:18<6:01:56,  1.86s/it]
 21%|██        | 3043/14725 [1:35:19<5:5
+0: {'loss': 0.8562, 'grad_norm': 0.9382491742929733, 'learning_rate': 2e-05, 'epoch': 0.21}
+0: 5:20,  1.83s/it]
 21%|██        | 3044/14725 [1:35:21<5:52:34,  1.81s/it]
 21%|██        | 3045/14725 [1:35:23<5:52:12,  1.81s/it]
 21%|██        | 3046/14725 [1:35:25<5:54:01,  1.82s/it]
 21%|██        | 3047/14725 [1:35:27<5:52:24,  1.81s/it]
 21%|██        | 3048/14725 [1:35:28<5:52:17,  1.81s/it]
 21%|██        | 3049/14725 [1:35:30<5:49:27,  1.80s/it]
 21%|██        | 3050/14725 [1:35:32<5:50:54,  1.80s/it]
                                                        

 21%|██        | 3050/14725 [1:35:32<5:50:54,  1.80s/it]
 21%|██        | 3051/14725 [1:35:34<5:50:06,  1.80s/it]
 21%|██        | 3052/14725 [1:35:36<5:48:52,  1.79s/it]
 21%|██        | 3053/14725 [1:35:37<5:49:16,  1.80s/it]
 21%|██        | 3054/14725 [1:35:39<5:51:43,  1.81s/it]
 21%|██        | 3055/14725 [1:35:41<5:49:38,  1.80s/it]
 21%|██        | 3056/14725 [1:35:43<5:53:26,  1.82s/it]
 21%|██        | 3057/14725 [1:35:45<5:54:47,  1.82s/it]
 21%|██        | 3058/14725 [1
+0: {'loss': 0.8508, 'grad_norm': 0.8404061616935858, 'learning_rate': 2e-05, 'epoch': 0.21}
+0: {'loss': 0.8783, 'grad_norm': 0.7825795270095073, 'learning_rate': 2e-05, 'epoch': 0.21}
+0: :35:47<5:52:38,  1.81s/it]
 21%|██        | 3059/14725 [1:35:48<5:54:03,  1.82s/it]
 21%|██        | 3060/14725 [1:35:50<5:55:11,  1.83s/it]
                                                        

 21%|██        | 3060/14725 [1:35:50<5:55:11,  1.83s/it]
 21%|██        | 3061/14725 [1:35:52<5:52:53,  1.82s/it]
 21%|██        | 3062/14725 [1:35:54<5:50:35,  1.80s/it]
 21%|██        | 3063/14725 [1:35:56<5:49:10,  1.80s/it]
 21%|██        | 3064/14725 [1:35:57<5:50:03,  1.80s/it]
 21%|██        | 3065/14725 [1:35:59<5:48:31,  1.79s/it]
 21%|██        | 3066/14725 [1:36:01<5:50:27,  1.80s/it]
 21%|██        | 3067/14725 [1:36:03<5:48:52,  1.80s/it]
 21%|██        | 3068/14725 [1:36:05<5:47:10,  1.79s/it]
 21%|██        | 3069/14725 [1:36:06<5:49:10,  1.80s/it]
 21%|██        | 3070/14725 [1:36:08<5:48:23,  1.79s/it]
                                                        

 21%|██        | 3070/14725 [1:36:08<5:48:23,  1.79s/it]
 21%|██        | 3071/1
+0: {'loss': 0.8679, 'grad_norm': 0.868211011909062, 'learning_rate': 2e-05, 'epoch': 0.21}
+0: 4725 [1:36:10<5:50:26,  1.80s/it]
 21%|██        | 3072/14725 [1:36:12<5:48:08,  1.79s/it]
 21%|██        | 3073/14725 [1:36:14<5:52:33,  1.82s/it]
 21%|██        | 3074/14725 [1:36:15<5:53:35,  1.82s/it]
 21%|██        | 3075/14725 [1:36:17<5:52:46,  1.82s/it]
 21%|██        | 3076/14725 [1:36:19<5:54:49,  1.83s/it]
 21%|██        | 3077/14725 [1:36:21<5:52:19,  1.81s/it]
 21%|██        | 3078/14725 [1:36:23<5:49:50,  1.80s/it]
 21%|██        | 3079/14725 [1:36:24<5:49:10,  1.80s/it]
 21%|██        | 3080/14725 [1:36:26<5:53:28,  1.82s/it]
                                                        

 21%|██        | 3080/14725 [1:36:26<5:53:28,  1.82s/it]
 21%|██        | 3081/14725 [1:36:28<5:52:15,  1.82s/it]
 21%|██        | 3082/14725 [1:36:30<5:54:48,  1.83s/it]
 21%|██        | 3083/14725 [1:36:32<5:50:59,  1.81s/it]
 21%|██        | 3084/14725 [1:36:34<5:54:13,  1.83s/it]
 21%|██        | 3085/14725 [1:36:35<5:56:32,  1.84s/it]
 21%|██      
+0: {'loss': 0.8788, 'grad_norm': 0.8250967953375697, 'learning_rate': 2e-05, 'epoch': 0.21}
+0:   | 3086/14725 [1:36:37<5:56:48,  1.84s/it]
 21%|██        | 3087/14725 [1:36:39<5:52:05,  1.82s/it]
 21%|██        | 3088/14725 [1:36:41<5:49:57,  1.80s/it]
 21%|██        | 3089/14725 [1:36:43<5:52:15,  1.82s/it]
 21%|██        | 3090/14725 [1:36:44<5:49:24,  1.80s/it]
                                                        

 21%|██        | 3090/14725 [1:36:44<5:49:24,  1.80s/it]
 21%|██        | 3091/14725 [1:36:46<5:48:07,  1.80s/it]
 21%|██        | 3092/14725 [1:36:48<5:52:47,  1.82s/it]
 21%|██        | 3093/14725 [1:36:50<5:50:32,  1.81s/it]
 21%|██        | 3094/14725 [1:36:52<5:47:39,  1.79s/it]
 21%|██        | 3095/14725 [1:36:54<5:52:29,  1.82s/it]
 21%|██        | 3096/14725 [1:36:55<5:50:23,  1.81s/it]
 21%|██        | 3097/14725 [1:36:57<5:50:24,  1.81s/it]
 21%|██        | 3098/14725 [1:36:59<5:50:50,  1.81s/it]
 21%|██        | 3099/14725 [1:37:01<5:50:40,  1.81s/it]
 21%|██        | 3100/14725 [1:37:03<5:48:45,  1.80s/it]
       
+0: {'loss': 0.87, 'grad_norm': 0.8145416738919966, 'learning_rate': 2e-05, 'epoch': 0.21}
+0: {'loss': 0.8713, 'grad_norm': 0.8052714029713054, 'learning_rate': 2e-05, 'epoch': 0.21}
+0:                                                  

 21%|██        | 3100/14725 [1:37:03<5:48:45,  1.80s/it]
 21%|██        | 3101/14725 [1:37:04<5:48:39,  1.80s/it]
 21%|██        | 3102/14725 [1:37:06<5:50:24,  1.81s/it]
 21%|██        | 3103/14725 [1:37:08<5:58:22,  1.85s/it]
 21%|██        | 3104/14725 [1:37:10<5:55:27,  1.84s/it]
 21%|██        | 3105/14725 [1:37:12<5:55:27,  1.84s/it]
 21%|██        | 3106/14725 [1:37:14<5:51:49,  1.82s/it]
 21%|██        | 3107/14725 [1:37:15<5:56:48,  1.84s/it]
 21%|██        | 3108/14725 [1:37:17<5:58:10,  1.85s/it]
 21%|██        | 3109/14725 [1:37:19<5:54:48,  1.83s/it]
 21%|██        | 3110/14725 [1:37:21<5:54:43,  1.83s/it]
                                                        

 21%|██        | 3110/14725 [1:37:21<5:54:43,  1.83s/it]
 21%|██        | 3111/14725 [1:37:23<5:54:58,  1.83s/it]
 21%|██        | 3112/14725 [1:37:25<5:51:22,  1.82s/it]
 21%|██        | 3113/14725 [1:37:26<5:51:42,  1.82s/it]
+0: {'loss': 0.8488, 'grad_norm': 0.7417129736794864, 'learning_rate': 2e-05, 'epoch': 0.21}
+0:  21%|██        | 3114/14725 [1:37:28<5:54:24,  1.83s/it]
 21%|██        | 3115/14725 [1:37:30<5:53:11,  1.83s/it]
 21%|██        | 3116/14725 [1:37:32<5:49:34,  1.81s/it]
 21%|██        | 3117/14725 [1:37:34<5:51:47,  1.82s/it]
 21%|██        | 3118/14725 [1:37:35<5:53:31,  1.83s/it]
 21%|██        | 3119/14725 [1:37:37<5:49:59,  1.81s/it]
 21%|██        | 3120/14725 [1:37:39<5:54:19,  1.83s/it]
                                                        

 21%|██        | 3120/14725 [1:37:39<5:54:19,  1.83s/it]
 21%|██        | 3121/14725 [1:37:41<5:58:36,  1.85s/it]
 21%|██        | 3122/14725 [1:37:43<5:59:24,  1.86s/it]
 21%|██        | 3123/14725 [1:37:45<5:57:03,  1.85s/it]
 21%|██        | 3124/14725 [1:37:46<5:52:51,  1.82s/it]
 21%|██        | 3125/14725 [1:37:48<5:49:10,  1.81s/it]
 21%|██        | 3126/14725 [1:37:50<5:51:52,  1.82s/it]
 21%|██        | 3127/14725 [1:37:52<5:51:01,  1.82s/it]
 21%|██        | 3128/14725 [1:37:54<5:52:30,  
+0: {'loss': 0.8686, 'grad_norm': 0.762814532419323, 'learning_rate': 2e-05, 'epoch': 0.21}
+0: {'loss': 0.857, 'grad_norm': 0.8468583731450274, 'learning_rate': 2e-05, 'epoch': 0.21}
+0: 1.82s/it]
 21%|██        | 3129/14725 [1:37:56<5:52:46,  1.83s/it]
 21%|██▏       | 3130/14725 [1:37:57<5:50:18,  1.81s/it]
                                                        

 21%|██▏       | 3130/14725 [1:37:57<5:50:18,  1.81s/it]
 21%|██▏       | 3131/14725 [1:37:59<5:51:30,  1.82s/it]
 21%|██▏       | 3132/14725 [1:38:01<5:47:50,  1.80s/it]
 21%|██▏       | 3133/14725 [1:38:03<5:47:46,  1.80s/it]
 21%|██▏       | 3134/14725 [1:38:05<5:47:42,  1.80s/it]
 21%|██▏       | 3135/14725 [1:38:07<6:10:48,  1.92s/it]
 21%|██▏       | 3136/14725 [1:38:09<6:05:13,  1.89s/it]
 21%|██▏       | 3137/14725 [1:38:11<6:07:01,  1.90s/it]
 21%|██▏       | 3138/14725 [1:38:12<6:08:45,  1.91s/it]
 21%|██▏       | 3139/14725 [1:38:14<6:05:17,  1.89s/it]
 21%|██▏       | 3140/14725 [1:38:16<6:02:30,  1.88s/it]
                                                        

 21%|██▏       | 3140/14725 [1:38:16<6:02:30,  1.88s/it]
 21%|██▏    
+0: {'loss': 0.8517, 'grad_norm': 0.8030210512693089, 'learning_rate': 2e-05, 'epoch': 0.21}
+0:    | 3141/14725 [1:38:18<6:00:54,  1.87s/it]
 21%|██▏       | 3142/14725 [1:38:20<5:58:01,  1.85s/it]
 21%|██▏       | 3143/14725 [1:38:22<5:58:51,  1.86s/it]
 21%|██▏       | 3144/14725 [1:38:23<5:55:28,  1.84s/it]
 21%|██▏       | 3145/14725 [1:38:25<5:55:44,  1.84s/it]
 21%|██▏       | 3146/14725 [1:38:27<5:58:48,  1.86s/it]
 21%|██▏       | 3147/14725 [1:38:29<5:55:30,  1.84s/it]
 21%|██▏       | 3148/14725 [1:38:31<5:51:22,  1.82s/it]
 21%|██▏       | 3149/14725 [1:38:33<5:49:21,  1.81s/it]
 21%|██▏       | 3150/14725 [1:38:34<5:49:02,  1.81s/it]
                                                        

 21%|██▏       | 3150/14725 [1:38:34<5:49:02,  1.81s/it]
 21%|██▏       | 3151/14725 [1:38:36<5:51:29,  1.82s/it]
 21%|██▏       | 3152/14725 [1:38:38<5:52:24,  1.83s/it]
 21%|██▏       | 3153/14725 [1:38:40<5:52:12,  1.83s/it]
 21%|██▏       | 3154/14725 [1:38:42<5:49:47,  1.81s/it]
 21%|██▏       | 3155/14725 [1:38
+0: {'loss': 0.8694, 'grad_norm': 0.8691597221377927, 'learning_rate': 2e-05, 'epoch': 0.21}
+0: :44<5:50:49,  1.82s/it]
 21%|██▏       | 3156/14725 [1:38:45<5:51:44,  1.82s/it]
 21%|██▏       | 3157/14725 [1:38:47<5:54:53,  1.84s/it]
 21%|██▏       | 3158/14725 [1:38:49<5:56:57,  1.85s/it]
 21%|██▏       | 3159/14725 [1:38:51<5:52:57,  1.83s/it]
 21%|██▏       | 3160/14725 [1:38:53<5:54:02,  1.84s/it]
                                                        

 21%|██▏       | 3160/14725 [1:38:53<5:54:02,  1.84s/it]
 21%|██▏       | 3161/14725 [1:38:55<5:54:37,  1.84s/it]
 21%|██▏       | 3162/14725 [1:38:56<5:51:45,  1.83s/it]
 21%|██▏       | 3163/14725 [1:38:58<5:56:23,  1.85s/it]
 21%|██▏       | 3164/14725 [1:39:00<5:52:39,  1.83s/it]
 21%|██▏       | 3165/14725 [1:39:02<5:50:20,  1.82s/it]
 22%|██▏       | 3166/14725 [1:39:04<5:53:10,  1.83s/it]
 22%|██▏       | 3167/14725 [1:39:05<5:49:09,  1.81s/it]
 22%|██▏       | 3168/14725 [1:39:07<5:46:42,  1.80s/it]
 22%|██▏       | 3169/14725 [1:39:09<5:45:16,  1.79s/i
+0: {'loss': 0.8649, 'grad_norm': 0.7622999675082428, 'learning_rate': 2e-05, 'epoch': 0.22}
+0: {'loss': 0.8452, 'grad_norm': 0.8104831892950698, 'learning_rate': 2e-05, 'epoch': 0.22}
+0: t]
 22%|██▏       | 3170/14725 [1:39:11<5:47:10,  1.80s/it]
                                                        

 22%|██▏       | 3170/14725 [1:39:11<5:47:10,  1.80s/it]
 22%|██▏       | 3171/14725 [1:39:13<5:45:15,  1.79s/it]
 22%|██▏       | 3172/14725 [1:39:14<5:44:38,  1.79s/it]
 22%|██▏       | 3173/14725 [1:39:16<5:46:34,  1.80s/it]
 22%|██▏       | 3174/14725 [1:39:18<5:45:22,  1.79s/it]
 22%|██▏       | 3175/14725 [1:39:20<5:45:34,  1.80s/it]
 22%|██▏       | 3176/14725 [1:39:22<5:48:32,  1.81s/it]
 22%|██▏       | 3177/14725 [1:39:23<5:47:20,  1.80s/it]
 22%|██▏       | 3178/14725 [1:39:25<5:49:53,  1.82s/it]
 22%|██▏       | 3179/14725 [1:39:27<5:46:35,  1.80s/it]
 22%|██▏       | 3180/14725 [1:39:29<5:45:36,  1.80s/it]
                                                        

 22%|██▏       | 3180/14725 [1:39:29<5:45:36,  1.80s/it]
 22%|██▏       | 3181/14725 [1:39:31<5:47:55,  1.81s/it]
 22%|██▏       | 
+0: {'loss': 0.8578, 'grad_norm': 0.809238272002198, 'learning_rate': 2e-05, 'epoch': 0.22}
+0: 3182/14725 [1:39:32<5:46:35,  1.80s/it]
 22%|██▏       | 3183/14725 [1:39:34<5:45:57,  1.80s/it]
 22%|██▏       | 3184/14725 [1:39:36<5:49:11,  1.82s/it]
 22%|██▏       | 3185/14725 [1:39:38<5:50:08,  1.82s/it]
 22%|██▏       | 3186/14725 [1:39:40<5:51:42,  1.83s/it]
 22%|██▏       | 3187/14725 [1:39:42<5:57:16,  1.86s/it]
 22%|██▏       | 3188/14725 [1:39:43<5:51:43,  1.83s/it]
 22%|██▏       | 3189/14725 [1:39:45<5:47:36,  1.81s/it]
 22%|██▏       | 3190/14725 [1:39:47<5:48:59,  1.82s/it]
                                                        

 22%|██▏       | 3190/14725 [1:39:47<5:48:59,  1.82s/it]
 22%|██▏       | 3191/14725 [1:39:49<5:54:34,  1.84s/it]
 22%|██▏       | 3192/14725 [1:39:51<5:50:22,  1.82s/it]
 22%|██▏       | 3193/14725 [1:39:53<5:46:57,  1.81s/it]
 22%|██▏       | 3194/14725 [1:39:54<5:48:21,  1.81s/it]
 22%|██▏       | 3195/14725 [1:39:56<5:50:15,  1.82s/it]
 22%|██▏       | 3196/14725 [1:39:58<5
+0: {'loss': 0.8778, 'grad_norm': 0.7765667872130849, 'learning_rate': 2e-05, 'epoch': 0.22}
+0: :49:38,  1.82s/it]
 22%|██▏       | 3197/14725 [1:40:00<5:46:54,  1.81s/it]
 22%|██▏       | 3198/14725 [1:40:02<5:44:31,  1.79s/it]
 22%|██▏       | 3199/14725 [1:40:03<5:45:35,  1.80s/it]
 22%|██▏       | 3200/14725 [1:40:05<5:44:16,  1.79s/it]
                                                        

 22%|██▏       | 3200/14725 [1:40:05<5:44:16,  1.79s/it]
 22%|██▏       | 3201/14725 [1:40:07<5:46:38,  1.80s/it]
 22%|██▏       | 3202/14725 [1:40:09<5:48:30,  1.81s/it]
 22%|██▏       | 3203/14725 [1:40:11<5:46:53,  1.81s/it]
 22%|██▏       | 3204/14725 [1:40:12<5:49:17,  1.82s/it]
 22%|██▏       | 3205/14725 [1:40:14<5:45:00,  1.80s/it]
 22%|██▏       | 3206/14725 [1:40:16<5:45:23,  1.80s/it]
 22%|██▏       | 3207/14725 [1:40:18<5:43:37,  1.79s/it]
 22%|██▏       | 3208/14725 [1:40:20<5:45:36,  1.80s/it]
 22%|██▏       | 3209/14725 [1:40:21<5:45:22,  1.80s/it]
 22%|██▏       | 3210/14725 [1:40:23<5:46:49,  1.81s/it]
  
+0: {'loss': 0.8651, 'grad_norm': 0.7589420416559447, 'learning_rate': 2e-05, 'epoch': 0.22}
+0: {'loss': 0.8578, 'grad_norm': 0.8120108354075897, 'learning_rate': 2e-05, 'epoch': 0.22}
+0:                                                       

 22%|██▏       | 3210/14725 [1:40:23<5:46:49,  1.81s/it]
 22%|██▏       | 3211/14725 [1:40:25<5:47:07,  1.81s/it]
 22%|██▏       | 3212/14725 [1:40:27<5:50:27,  1.83s/it]
 22%|██▏       | 3213/14725 [1:40:29<5:52:00,  1.83s/it]
 22%|██▏       | 3214/14725 [1:40:31<5:53:19,  1.84s/it]
 22%|██▏       | 3215/14725 [1:40:32<5:50:54,  1.83s/it]
 22%|██▏       | 3216/14725 [1:40:34<5:47:24,  1.81s/it]
 22%|██▏       | 3217/14725 [1:40:36<5:49:27,  1.82s/it]
 22%|██▏       | 3218/14725 [1:40:38<5:52:41,  1.84s/it]
 22%|██▏       | 3219/14725 [1:40:40<5:53:22,  1.84s/it]
 22%|██▏       | 3220/14725 [1:40:42<5:49:08,  1.82s/it]
                                                        

 22%|██▏       | 3220/14725 [1:40:42<5:49:08,  1.82s/it]
 22%|██▏       | 3221/14725 [1:40:43<5:50:55,  1.83s/it]
 22%|██▏       | 3222/14725 [1:40:45<5:50:27,  1.83s/it]
 22%|██▏       | 3223/
+0: {'loss': 0.8509, 'grad_norm': 0.8562610155204394, 'learning_rate': 2e-05, 'epoch': 0.22}
+0: 14725 [1:40:47<5:49:14,  1.82s/it]
 22%|██▏       | 3224/14725 [1:40:49<5:47:16,  1.81s/it]
 22%|██▏       | 3225/14725 [1:40:51<5:45:58,  1.81s/it]
 22%|██▏       | 3226/14725 [1:40:52<5:48:14,  1.82s/it]
 22%|██▏       | 3227/14725 [1:40:54<5:50:32,  1.83s/it]
 22%|██▏       | 3228/14725 [1:40:56<5:46:39,  1.81s/it]
 22%|██▏       | 3229/14725 [1:40:58<5:45:36,  1.80s/it]
 22%|██▏       | 3230/14725 [1:41:00<5:45:07,  1.80s/it]
                                                        

 22%|██▏       | 3230/14725 [1:41:00<5:45:07,  1.80s/it]
 22%|██▏       | 3231/14725 [1:41:01<5:43:00,  1.79s/it]
 22%|██▏       | 3232/14725 [1:41:03<5:44:46,  1.80s/it]
 22%|██▏       | 3233/14725 [1:41:05<5:44:32,  1.80s/it]
 22%|██▏       | 3234/14725 [1:41:07<5:44:44,  1.80s/it]
 22%|██▏       | 3235/14725 [1:41:09<5:43:44,  1.79s/it]
 22%|██▏       | 3236/14725 [1:41:10<5:46:14,  1.81s/it]
 22%|██▏       | 3237/14725 [1:41:12<5:47:3
+0: {'loss': 0.8493, 'grad_norm': 0.9108596862057341, 'learning_rate': 2e-05, 'epoch': 0.22}
+0: {'loss': 0.8451, 'grad_norm': 0.7894185236877271, 'learning_rate': 2e-05, 'epoch': 0.22}
+0: 5,  1.82s/it]
 22%|██▏       | 3238/14725 [1:41:14<5:44:10,  1.80s/it]
 22%|██▏       | 3239/14725 [1:41:16<5:45:14,  1.80s/it]
 22%|██▏       | 3240/14725 [1:41:18<5:45:13,  1.80s/it]
                                                        

 22%|██▏       | 3240/14725 [1:41:18<5:45:13,  1.80s/it]
 22%|██▏       | 3241/14725 [1:41:20<5:48:37,  1.82s/it]
 22%|██▏       | 3242/14725 [1:41:21<5:45:11,  1.80s/it]
 22%|██▏       | 3243/14725 [1:41:23<5:44:27,  1.80s/it]
 22%|██▏       | 3244/14725 [1:41:25<5:45:20,  1.80s/it]
 22%|██▏       | 3245/14725 [1:41:27<5:48:51,  1.82s/it]
 22%|██▏       | 3246/14725 [1:41:29<5:45:48,  1.81s/it]
 22%|██▏       | 3247/14725 [1:41:30<5:45:30,  1.81s/it]
 22%|██▏       | 3248/14725 [1:41:32<5:48:55,  1.82s/it]
 22%|██▏       | 3249/14725 [1:41:34<5:47:20,  1.82s/it]
 22%|██▏       | 3250/14725 [1:41:36<5:49:48,  1.83s/it]
                                                        

 22%|██�
+0: {'loss': 0.8745, 'grad_norm': 0.7841609773641796, 'learning_rate': 2e-05, 'epoch': 0.22}
+0: ��       | 3250/14725 [1:41:36<5:49:48,  1.83s/it]
 22%|██▏       | 3251/14725 [1:41:38<5:48:25,  1.82s/it]
 22%|██▏       | 3252/14725 [1:41:39<5:46:07,  1.81s/it]
 22%|██▏       | 3253/14725 [1:41:41<5:47:00,  1.81s/it]
 22%|██▏       | 3254/14725 [1:41:43<5:44:01,  1.80s/it]
 22%|██▏       | 3255/14725 [1:41:45<5:43:45,  1.80s/it]
 22%|██▏       | 3256/14725 [1:41:47<5:44:04,  1.80s/it]
 22%|██▏       | 3257/14725 [1:41:48<5:44:27,  1.80s/it]
 22%|██▏       | 3258/14725 [1:41:50<5:45:17,  1.81s/it]
 22%|██▏       | 3259/14725 [1:41:52<5:45:17,  1.81s/it]
 22%|██▏       | 3260/14725 [1:41:54<5:50:10,  1.83s/it]
                                                        

 22%|██▏       | 3260/14725 [1:41:54<5:50:10,  1.83s/it]
 22%|██▏       | 3261/14725 [1:41:56<5:51:07,  1.84s/it]
 22%|██▏       | 3262/14725 [1:41:58<5:53:31,  1.85s/it]
 22%|██▏       | 3263/14725 [1:42:00<5:51:25,  1.84s/it]
 22%|██▏       | 3264/14725
+0: {'loss': 0.8545, 'grad_norm': 0.7305809126507787, 'learning_rate': 2e-05, 'epoch': 0.22}
+0:  [1:42:01<5:50:28,  1.83s/it]
 22%|██▏       | 3265/14725 [1:42:03<5:46:07,  1.81s/it]
 22%|██▏       | 3266/14725 [1:42:05<5:49:13,  1.83s/it]
 22%|██▏       | 3267/14725 [1:42:07<5:53:16,  1.85s/it]
 22%|██▏       | 3268/14725 [1:42:09<5:54:03,  1.85s/it]
 22%|██▏       | 3269/14725 [1:42:11<5:54:22,  1.86s/it]
 22%|██▏       | 3270/14725 [1:42:12<5:48:06,  1.82s/it]
                                                        

 22%|██▏       | 3270/14725 [1:42:12<5:48:06,  1.82s/it]
 22%|██▏       | 3271/14725 [1:42:14<5:48:47,  1.83s/it]
 22%|██▏       | 3272/14725 [1:42:16<5:45:11,  1.81s/it]
 22%|██▏       | 3273/14725 [1:42:18<5:48:00,  1.82s/it]
 22%|██▏       | 3274/14725 [1:42:20<5:46:12,  1.81s/it]
 22%|██▏       | 3275/14725 [1:42:21<5:44:46,  1.81s/it]
 22%|██▏       | 3276/14725 [1:42:23<5:42:09,  1.79s/it]
 22%|██▏       | 3277/14725 [1:42:25<5:43:35,  1.80s/it]
 22%|██▏       | 3278/14725 [1:42:27<5:42:33,  1
+0: {'loss': 0.8666, 'grad_norm': 0.843509592444824, 'learning_rate': 2e-05, 'epoch': 0.22}
+0: {'loss': 0.8483, 'grad_norm': 0.7892030532170013, 'learning_rate': 2e-05, 'epoch': 0.22}
+0: .80s/it]
 22%|██▏       | 3279/14725 [1:42:29<5:43:13,  1.80s/it]
 22%|██▏       | 3280/14725 [1:42:30<5:41:02,  1.79s/it]
                                                        

 22%|██▏       | 3280/14725 [1:42:30<5:41:02,  1.79s/it]
 22%|██▏       | 3281/14725 [1:42:32<5:42:23,  1.80s/it]
 22%|██▏       | 3282/14725 [1:42:34<5:44:47,  1.81s/it]
 22%|██▏       | 3283/14725 [1:42:36<5:42:50,  1.80s/it]
 22%|██▏       | 3284/14725 [1:42:38<5:45:26,  1.81s/it]
 22%|██▏       | 3285/14725 [1:42:39<5:50:53,  1.84s/it]
 22%|██▏       | 3286/14725 [1:42:41<5:46:06,  1.82s/it]
 22%|██▏       | 3287/14725 [1:42:43<5:43:54,  1.80s/it]
 22%|██▏       | 3288/14725 [1:42:45<5:41:02,  1.79s/it]
 22%|██▏       | 3289/14725 [1:42:47<5:48:50,  1.83s/it]
 22%|██▏       | 3290/14725 [1:42:49<5:50:23,  1.84s/it]
                                                        

 22%|██▏       | 3290/14725 [1:42:49<5:50:23,  1.84s/it]
 22%|██▏   
+0: {'loss': 0.8743, 'grad_norm': 0.8116508124766203, 'learning_rate': 2e-05, 'epoch': 0.22}
+0:     | 3291/14725 [1:42:50<5:47:26,  1.82s/it]
 22%|██▏       | 3292/14725 [1:42:52<5:43:31,  1.80s/it]
 22%|██▏       | 3293/14725 [1:42:54<5:43:49,  1.80s/it]
 22%|██▏       | 3294/14725 [1:42:56<5:42:56,  1.80s/it]
 22%|██▏       | 3295/14725 [1:42:58<5:44:39,  1.81s/it]
 22%|██▏       | 3296/14725 [1:42:59<5:45:57,  1.82s/it]
 22%|██▏       | 3297/14725 [1:43:01<5:47:24,  1.82s/it]
 22%|██▏       | 3298/14725 [1:43:03<5:46:16,  1.82s/it]
 22%|██▏       | 3299/14725 [1:43:05<5:49:11,  1.83s/it]
 22%|██▏       | 3300/14725 [1:43:07<5:49:58,  1.84s/it]
                                                        

 22%|██▏       | 3300/14725 [1:43:07<5:49:58,  1.84s/it]
 22%|██▏       | 3301/14725 [1:43:09<5:49:49,  1.84s/it]
 22%|██▏       | 3302/14725 [1:43:10<5:46:04,  1.82s/it]
 22%|██▏       | 3303/14725 [1:43:12<5:46:49,  1.82s/it]
 22%|██▏       | 3304/14725 [1:43:14<5:43:01,  1.80s/it]
 22%|██▏       | 3305/14725 [1:4
+0: {'loss': 0.8461, 'grad_norm': 0.8326953042765947, 'learning_rate': 2e-05, 'epoch': 0.22}
+0: 3:16<5:40:41,  1.79s/it]
 22%|██▏       | 3306/14725 [1:43:18<5:45:51,  1.82s/it]
 22%|██▏       | 3307/14725 [1:43:19<5:43:32,  1.81s/it]
 22%|██▏       | 3308/14725 [1:43:21<5:42:39,  1.80s/it]
 22%|██▏       | 3309/14725 [1:43:23<5:46:36,  1.82s/it]
 22%|██▏       | 3310/14725 [1:43:25<5:46:33,  1.82s/it]
                                                        

 22%|██▏       | 3310/14725 [1:43:25<5:46:33,  1.82s/it]
 22%|██▏       | 3311/14725 [1:43:27<5:44:19,  1.81s/it]
 22%|██▏       | 3312/14725 [1:43:28<5:49:15,  1.84s/it]
 22%|██▏       | 3313/14725 [1:43:30<5:47:47,  1.83s/it]
 23%|██▎       | 3314/14725 [1:43:32<5:53:02,  1.86s/it]
 23%|██▎       | 3315/14725 [1:43:34<5:47:55,  1.83s/it]
 23%|██▎       | 3316/14725 [1:43:36<5:43:50,  1.81s/it]
 23%|██▎       | 3317/14725 [1:43:38<5:42:02,  1.80s/it]
 23%|██▎       | 3318/14725 [1:43:39<5:42:41,  1.80s/it]
 23%|██▎       | 3319/14725 [1:43:41<5:43:44,  1.81s/
+0: {'loss': 0.8663, 'grad_norm': 0.8453082694463709, 'learning_rate': 2e-05, 'epoch': 0.23}
+0: {'loss': 0.852, 'grad_norm': 0.8148205898766292, 'learning_rate': 2e-05, 'epoch': 0.23}
+0: it]
 23%|██▎       | 3320/14725 [1:43:43<5:41:39,  1.80s/it]
                                                        

 23%|██▎       | 3320/14725 [1:43:43<5:41:39,  1.80s/it]
 23%|██▎       | 3321/14725 [1:43:45<5:41:12,  1.80s/it]
 23%|██▎       | 3322/14725 [1:43:46<5:40:04,  1.79s/it]
 23%|██▎       | 3323/14725 [1:43:48<5:40:42,  1.79s/it]
 23%|██▎       | 3324/14725 [1:43:50<5:39:23,  1.79s/it]
 23%|██▎       | 3325/14725 [1:43:52<5:38:40,  1.78s/it]
 23%|██▎       | 3326/14725 [1:43:54<5:38:26,  1.78s/it]
 23%|██▎       | 3327/14725 [1:43:55<5:42:49,  1.80s/it]
 23%|██▎       | 3328/14725 [1:43:57<5:40:48,  1.79s/it]
 23%|██▎       | 3329/14725 [1:43:59<5:41:16,  1.80s/it]
 23%|██▎       | 3330/14725 [1:44:01<5:39:07,  1.79s/it]
                                                        

 23%|██▎       | 3330/14725 [1:44:01<5:39:07,  1.79s/it]
 23%|██▎       | 3331/14725 [1:44:03<5:42:06,  1.80s/it]
 23%|██▎       |
+0: {'loss': 0.8602, 'grad_norm': 0.849159092556597, 'learning_rate': 2e-05, 'epoch': 0.23}
+0:  3332/14725 [1:44:05<5:45:11,  1.82s/it]
 23%|██▎       | 3333/14725 [1:44:06<5:44:42,  1.82s/it]
 23%|██▎       | 3334/14725 [1:44:08<5:40:37,  1.79s/it]
 23%|██▎       | 3335/14725 [1:44:10<5:38:29,  1.78s/it]
 23%|██▎       | 3336/14725 [1:44:12<5:38:29,  1.78s/it]
 23%|██▎       | 3337/14725 [1:44:13<5:36:52,  1.77s/it]
 23%|██▎       | 3338/14725 [1:44:15<5:35:25,  1.77s/it]
 23%|██▎       | 3339/14725 [1:44:17<5:36:39,  1.77s/it]
 23%|██▎       | 3340/14725 [1:44:19<5:38:06,  1.78s/it]
                                                        

 23%|██▎       | 3340/14725 [1:44:19<5:38:06,  1.78s/it]
 23%|██▎       | 3341/14725 [1:44:20<5:37:31,  1.78s/it]
 23%|██▎       | 3342/14725 [1:44:22<5:40:30,  1.79s/it]
 23%|██▎       | 3343/14725 [1:44:24<5:42:04,  1.80s/it]
 23%|██▎       | 3344/14725 [1:44:26<5:38:56,  1.79s/it]
 23%|██▎       | 3345/14725 [1:44:28<5:40:58,  1.80s/it]
 23%|██▎       | 3346/14725 [1:44:30<
+0: {'loss': 0.8512, 'grad_norm': 0.8615025091730558, 'learning_rate': 2e-05, 'epoch': 0.23}
+0: 5:42:33,  1.81s/it]
 23%|██▎       | 3347/14725 [1:44:31<5:45:15,  1.82s/it]
 23%|██▎       | 3348/14725 [1:44:33<5:42:56,  1.81s/it]
 23%|██▎       | 3349/14725 [1:44:35<5:44:39,  1.82s/it]
 23%|██▎       | 3350/14725 [1:44:37<5:42:29,  1.81s/it]
                                                        

 23%|██▎       | 3350/14725 [1:44:37<5:42:29,  1.81s/it]
 23%|██▎       | 3351/14725 [1:44:39<5:40:52,  1.80s/it]
 23%|██▎       | 3352/14725 [1:44:40<5:40:48,  1.80s/it]
 23%|██▎       | 3353/14725 [1:44:42<5:42:23,  1.81s/it]
 23%|██▎       | 3354/14725 [1:44:44<5:42:53,  1.81s/it]
 23%|██▎       | 3355/14725 [1:44:46<5:42:56,  1.81s/it]
 23%|██▎       | 3356/14725 [1:44:48<5:47:30,  1.83s/it]
 23%|██▎       | 3357/14725 [1:44:50<5:46:17,  1.83s/it]
 23%|██▎       | 3358/14725 [1:44:51<5:42:14,  1.81s/it]
 23%|██▎       | 3359/14725 [1:44:53<5:43:05,  1.81s/it]
 23%|██▎       | 3360/14725 [1:44:55<5:46:46,  1.83s/it]
 
+0: {'loss': 0.8771, 'grad_norm': 0.7606797261541771, 'learning_rate': 2e-05, 'epoch': 0.23}
+0: {'loss': 0.8661, 'grad_norm': 0.8219713261920937, 'learning_rate': 2e-05, 'epoch': 0.23}
+0:                                                        

 23%|██▎       | 3360/14725 [1:44:55<5:46:46,  1.83s/it]
 23%|██▎       | 3361/14725 [1:44:57<5:43:50,  1.82s/it]
 23%|██▎       | 3362/14725 [1:44:59<5:41:24,  1.80s/it]
 23%|██▎       | 3363/14725 [1:45:00<5:43:54,  1.82s/it]
 23%|██▎       | 3364/14725 [1:45:02<5:41:47,  1.81s/it]
 23%|██▎       | 3365/14725 [1:45:04<5:38:37,  1.79s/it]
 23%|██▎       | 3366/14725 [1:45:06<5:36:14,  1.78s/it]
 23%|██▎       | 3367/14725 [1:45:07<5:39:40,  1.79s/it]
 23%|██▎       | 3368/14725 [1:45:09<5:40:05,  1.80s/it]
 23%|██▎       | 3369/14725 [1:45:11<5:42:55,  1.81s/it]
 23%|██▎       | 3370/14725 [1:45:13<5:41:12,  1.80s/it]
                                                        

 23%|██▎       | 3370/14725 [1:45:13<5:41:12,  1.80s/it]
 23%|██▎       | 3371/14725 [1:45:15<5:38:46,  1.79s/it]
 23%|██▎       | 3372/14725 [1:45:16<5:41:08,  1.80s/it]
 23%|██▎       | 3373
+0: {'loss': 0.8652, 'grad_norm': 0.8277480751383153, 'learning_rate': 2e-05, 'epoch': 0.23}
+0: /14725 [1:45:18<5:41:02,  1.80s/it]
 23%|██▎       | 3374/14725 [1:45:20<5:41:10,  1.80s/it]
 23%|██▎       | 3375/14725 [1:45:22<5:42:18,  1.81s/it]
 23%|██▎       | 3376/14725 [1:45:24<5:40:23,  1.80s/it]
 23%|██▎       | 3377/14725 [1:45:25<5:38:58,  1.79s/it]
 23%|██▎       | 3378/14725 [1:45:27<5:41:30,  1.81s/it]
 23%|██▎       | 3379/14725 [1:45:29<5:43:59,  1.82s/it]
 23%|██▎       | 3380/14725 [1:45:31<5:46:29,  1.83s/it]
                                                        

 23%|██▎       | 3380/14725 [1:45:31<5:46:29,  1.83s/it]
 23%|██▎       | 3381/14725 [1:45:33<5:49:49,  1.85s/it]
 23%|██▎       | 3382/14725 [1:45:35<5:49:49,  1.85s/it]
 23%|██▎       | 3383/14725 [1:45:37<5:49:26,  1.85s/it]
 23%|██▎       | 3384/14725 [1:45:38<5:47:06,  1.84s/it]
 23%|██▎       | 3385/14725 [1:45:40<5:44:08,  1.82s/it]
 23%|██▎       | 3386/14725 [1:45:42<5:42:06,  1.81s/it]
 23%|██▎       | 3387/14725 [1:45:44<5:41:
+0: {'loss': 0.8462, 'grad_norm': 0.7846290551223958, 'learning_rate': 2e-05, 'epoch': 0.23}
+0: {'loss': 0.8712, 'grad_norm': 0.9550338642041815, 'learning_rate': 2e-05, 'epoch': 0.23}
+0: 26,  1.81s/it]
 23%|██▎       | 3388/14725 [1:45:46<5:44:51,  1.83s/it]
 23%|██▎       | 3389/14725 [1:45:48<5:48:46,  1.85s/it]
 23%|██▎       | 3390/14725 [1:45:49<5:43:35,  1.82s/it]
                                                        

 23%|██▎       | 3390/14725 [1:45:49<5:43:35,  1.82s/it]
 23%|██▎       | 3391/14725 [1:45:51<5:46:16,  1.83s/it]
 23%|██▎       | 3392/14725 [1:45:53<5:46:39,  1.84s/it]
 23%|██▎       | 3393/14725 [1:45:55<5:51:55,  1.86s/it]
 23%|██▎       | 3394/14725 [1:45:57<5:49:15,  1.85s/it]
 23%|██▎       | 3395/14725 [1:45:59<5:51:01,  1.86s/it]
 23%|██▎       | 3396/14725 [1:46:00<5:48:37,  1.85s/it]
 23%|██▎       | 3397/14725 [1:46:02<5:44:10,  1.82s/it]
 23%|██▎       | 3398/14725 [1:46:04<5:42:57,  1.82s/it]
 23%|██▎       | 3399/14725 [1:46:06<5:40:22,  1.80s/it]
 23%|██▎       | 3400/14725 [1:46:08<5:45:08,  1.83s/it]
                                                        

 23%|██
+0: {'loss': 0.8545, 'grad_norm': 0.7743990525898311, 'learning_rate': 2e-05, 'epoch': 0.23}
+0: ▎       | 3400/14725 [1:46:08<5:45:08,  1.83s/it]
 23%|██▎       | 3401/14725 [1:46:10<5:46:40,  1.84s/it]
 23%|██▎       | 3402/14725 [1:46:11<5:43:35,  1.82s/it]
 23%|██▎       | 3403/14725 [1:46:13<5:43:54,  1.82s/it]
 23%|██▎       | 3404/14725 [1:46:15<5:42:42,  1.82s/it]
 23%|██▎       | 3405/14725 [1:46:17<5:44:32,  1.83s/it]
 23%|██▎       | 3406/14725 [1:46:19<5:42:09,  1.81s/it]
 23%|██▎       | 3407/14725 [1:46:20<5:39:20,  1.80s/it]
 23%|██▎       | 3408/14725 [1:46:22<5:43:31,  1.82s/it]
 23%|██▎       | 3409/14725 [1:46:24<5:45:22,  1.83s/it]
 23%|██▎       | 3410/14725 [1:46:26<5:43:44,  1.82s/it]
                                                        

 23%|██▎       | 3410/14725 [1:46:26<5:43:44,  1.82s/it]
 23%|██▎       | 3411/14725 [1:46:28<5:48:49,  1.85s/it]
 23%|██▎       | 3412/14725 [1:46:30<5:48:31,  1.85s/it]
 23%|██▎       | 3413/14725 [1:46:31<5:43:52,  1.82s/it]
 23%|██▎       | 3414/1472
+0: {'loss': 0.8589, 'grad_norm': 0.8503036545463469, 'learning_rate': 2e-05, 'epoch': 0.23}
+0: 5 [1:46:33<5:42:45,  1.82s/it]
 23%|██▎       | 3415/14725 [1:46:35<5:39:07,  1.80s/it]
 23%|██▎       | 3416/14725 [1:46:37<5:42:37,  1.82s/it]
 23%|██▎       | 3417/14725 [1:46:39<5:42:06,  1.82s/it]
 23%|██▎       | 3418/14725 [1:46:41<5:44:58,  1.83s/it]
 23%|██▎       | 3419/14725 [1:46:42<5:44:08,  1.83s/it]
 23%|██▎       | 3420/14725 [1:46:44<5:41:10,  1.81s/it]
                                                        

 23%|██▎       | 3420/14725 [1:46:44<5:41:10,  1.81s/it]
 23%|██▎       | 3421/14725 [1:46:46<5:43:38,  1.82s/it]
 23%|██▎       | 3422/14725 [1:46:48<5:41:04,  1.81s/it]
 23%|██▎       | 3423/14725 [1:46:50<5:41:08,  1.81s/it]
 23%|██▎       | 3424/14725 [1:46:52<6:02:33,  1.92s/it]
 23%|██▎       | 3425/14725 [1:46:54<5:54:43,  1.88s/it]
 23%|██▎       | 3426/14725 [1:46:55<5:50:01,  1.86s/it]
 23%|██▎       | 3427/14725 [1:46:57<5:44:45,  1.83s/it]
 23%|██▎       | 3428/14725 [1:46:59<5:41:14,  
+0: {'loss': 0.8625, 'grad_norm': 0.8580367512639213, 'learning_rate': 2e-05, 'epoch': 0.23}
+0: {'loss': 0.8665, 'grad_norm': 0.7890946774652389, 'learning_rate': 2e-05, 'epoch': 0.23}
+0: 1.81s/it]
 23%|██▎       | 3429/14725 [1:47:01<6:10:47,  1.97s/it]
 23%|██▎       | 3430/14725 [1:47:03<6:28:14,  2.06s/it]
                                                        

 23%|██▎       | 3430/14725 [1:47:03<6:28:14,  2.06s/it]
 23%|██▎       | 3431/14725 [1:47:05<6:13:24,  1.98s/it]
 23%|██▎       | 3432/14725 [1:47:07<6:02:39,  1.93s/it]
 23%|██▎       | 3433/14725 [1:47:09<5:52:33,  1.87s/it]
 23%|██▎       | 3434/14725 [1:47:11<5:47:58,  1.85s/it]
 23%|██▎       | 3435/14725 [1:47:12<5:43:49,  1.83s/it]
 23%|██▎       | 3436/14725 [1:47:14<5:44:42,  1.83s/it]
 23%|██▎       | 3437/14725 [1:47:16<5:39:30,  1.80s/it]
 23%|██▎       | 3438/14725 [1:47:18<5:41:49,  1.82s/it]
 23%|██▎       | 3439/14725 [1:47:20<6:09:18,  1.96s/it]
 23%|██▎       | 3440/14725 [1:47:22<5:59:19,  1.91s/it]
                                                        

 23%|██▎       | 3440/14725 [1:47:22<5:59:19,  1.91s/it]
 23%|██▎  
+0: {'loss': 0.8534, 'grad_norm': 0.7784174641556073, 'learning_rate': 2e-05, 'epoch': 0.23}
+0:      | 3441/14725 [1:47:24<5:57:16,  1.90s/it]
 23%|██▎       | 3442/14725 [1:47:26<5:53:42,  1.88s/it]
 23%|██▎       | 3443/14725 [1:47:27<5:49:34,  1.86s/it]
 23%|██▎       | 3444/14725 [1:47:29<5:43:19,  1.83s/it]
 23%|██▎       | 3445/14725 [1:47:31<5:43:37,  1.83s/it]
 23%|██▎       | 3446/14725 [1:47:33<5:41:19,  1.82s/it]
 23%|██▎       | 3447/14725 [1:47:35<5:40:13,  1.81s/it]
 23%|██▎       | 3448/14725 [1:47:36<5:42:19,  1.82s/it]
 23%|██▎       | 3449/14725 [1:47:38<5:38:44,  1.80s/it]
 23%|██▎       | 3450/14725 [1:47:40<5:37:12,  1.79s/it]
                                                        

 23%|██▎       | 3450/14725 [1:47:40<5:37:12,  1.79s/it]
 23%|██▎       | 3451/14725 [1:47:42<5:37:02,  1.79s/it]
 23%|██▎       | 3452/14725 [1:47:44<5:40:57,  1.81s/it]
 23%|██▎       | 3453/14725 [1:47:45<5:42:08,  1.82s/it]
 23%|██▎       | 3454/14725 [1:47:47<5:42:06,  1.82s/it]
 23%|██▎       | 3455/14725 [1:
+0: {'loss': 0.8597, 'grad_norm': 0.7761611844503916, 'learning_rate': 2e-05, 'epoch': 0.23}
+0: 47:49<5:40:00,  1.81s/it]
 23%|██▎       | 3456/14725 [1:47:51<6:05:05,  1.94s/it]
 23%|██▎       | 3457/14725 [1:47:53<5:59:09,  1.91s/it]
 23%|██▎       | 3458/14725 [1:47:55<5:58:12,  1.91s/it]
 23%|██▎       | 3459/14725 [1:47:57<5:53:58,  1.89s/it]
 23%|██▎       | 3460/14725 [1:47:59<5:53:13,  1.88s/it]
                                                        

 23%|██▎       | 3460/14725 [1:47:59<5:53:13,  1.88s/it]
 24%|██▎       | 3461/14725 [1:48:01<5:53:07,  1.88s/it]
 24%|██▎       | 3462/14725 [1:48:02<5:47:08,  1.85s/it]
 24%|██▎       | 3463/14725 [1:48:04<5:45:34,  1.84s/it]
 24%|██▎       | 3464/14725 [1:48:06<5:46:31,  1.85s/it]
 24%|██▎       | 3465/14725 [1:48:08<5:45:32,  1.84s/it]
 24%|██▎       | 3466/14725 [1:48:10<5:42:23,  1.82s/it]
 24%|██▎       | 3467/14725 [1:48:11<5:38:20,  1.80s/it]
 24%|██▎       | 3468/14725 [1:48:13<5:45:22,  1.84s/it]
 24%|██▎       | 3469/14725 [1:48:15<5:43:24,  1.83s
+0: {'loss': 0.8388, 'grad_norm': 0.7468271995234665, 'learning_rate': 2e-05, 'epoch': 0.24}
+0: {'loss': 0.8641, 'grad_norm': 0.7706317195406225, 'learning_rate': 2e-05, 'epoch': 0.24}
+0: /it]
 24%|██▎       | 3470/14725 [1:48:17<5:40:37,  1.82s/it]
                                                        

 24%|██▎       | 3470/14725 [1:48:17<5:40:37,  1.82s/it]
 24%|██▎       | 3471/14725 [1:48:19<5:42:51,  1.83s/it]
 24%|██▎       | 3472/14725 [1:48:21<5:45:25,  1.84s/it]
 24%|██▎       | 3473/14725 [1:48:22<5:40:54,  1.82s/it]
 24%|██▎       | 3474/14725 [1:48:24<5:39:35,  1.81s/it]
 24%|██▎       | 3475/14725 [1:48:26<5:38:37,  1.81s/it]
 24%|██▎       | 3476/14725 [1:48:28<5:39:08,  1.81s/it]
 24%|██▎       | 3477/14725 [1:48:30<5:44:28,  1.84s/it]
 24%|██▎       | 3478/14725 [1:48:32<5:47:08,  1.85s/it]
 24%|██▎       | 3479/14725 [1:48:34<5:45:45,  1.84s/it]
 24%|██▎       | 3480/14725 [1:48:35<5:44:54,  1.84s/it]
                                                        

 24%|██▎       | 3480/14725 [1:48:35<5:44:54,  1.84s/it]
 24%|██▎       | 3481/14725 [1:48:37<5:42:07,  1.83s/it]
 24%|██▎       
+0: {'loss': 0.8476, 'grad_norm': 0.8026897408751665, 'learning_rate': 2e-05, 'epoch': 0.24}
+0: | 3482/14725 [1:48:39<5:45:05,  1.84s/it]
 24%|██▎       | 3483/14725 [1:48:41<5:53:15,  1.89s/it]
 24%|██▎       | 3484/14725 [1:48:43<5:50:14,  1.87s/it]
 24%|██▎       | 3485/14725 [1:48:45<5:45:04,  1.84s/it]
 24%|██▎       | 3486/14725 [1:48:47<5:48:57,  1.86s/it]
 24%|██▎       | 3487/14725 [1:48:48<5:47:30,  1.86s/it]
 24%|██▎       | 3488/14725 [1:48:50<5:51:00,  1.87s/it]
 24%|██▎       | 3489/14725 [1:48:52<5:48:01,  1.86s/it]
 24%|██▎       | 3490/14725 [1:48:54<5:51:53,  1.88s/it]
                                                        

 24%|██▎       | 3490/14725 [1:48:54<5:51:53,  1.88s/it]
 24%|██▎       | 3491/14725 [1:48:56<5:51:44,  1.88s/it]
 24%|██▎       | 3492/14725 [1:48:58<5:50:15,  1.87s/it]
 24%|██▎       | 3493/14725 [1:49:00<5:47:16,  1.86s/it]
 24%|██▎       | 3494/14725 [1:49:01<5:43:39,  1.84s/it]
 24%|██▎       | 3495/14725 [1:49:03<5:47:22,  1.86s/it]
 24%|██▎       | 3496/14725 [1:49:05
+0: {'loss': 0.882, 'grad_norm': 0.7645733563144749, 'learning_rate': 2e-05, 'epoch': 0.24}
+0: <5:45:31,  1.85s/it]
 24%|██▎       | 3497/14725 [1:49:07<5:43:52,  1.84s/it]
 24%|██▍       | 3498/14725 [1:49:09<5:39:21,  1.81s/it]
 24%|██▍       | 3499/14725 [1:49:10<5:35:38,  1.79s/it]
 24%|██▍       | 3500/14725 [1:49:12<5:41:16,  1.82s/it]
                                                        

 24%|██▍       | 3500/14725 [1:49:12<5:41:16,  1.82s/it]
 24%|██▍       | 3501/14725 [1:49:14<5:41:36,  1.83s/it]
 24%|██▍       | 3502/14725 [1:49:16<5:42:27,  1.83s/it]
 24%|██▍       | 3503/14725 [1:49:18<5:45:41,  1.85s/it]
 24%|██▍       | 3504/14725 [1:49:20<5:42:06,  1.83s/it]
 24%|██▍       | 3505/14725 [1:49:21<5:37:47,  1.81s/it]
 24%|██▍       | 3506/14725 [1:49:23<5:41:20,  1.83s/it]
 24%|██▍       | 3507/14725 [1:49:25<5:41:44,  1.83s/it]
 24%|██▍       | 3508/14725 [1:49:27<5:39:07,  1.81s/it]
 24%|██▍       | 3509/14725 [1:49:29<5:36:24,  1.80s/it]
 24%|██▍       | 3510/14725 [1:49:30<5:36:28,  1.80s/it]
+0: {'loss': 0.8564, 'grad_norm': 0.7954160081267244, 'learning_rate': 2e-05, 'epoch': 0.24}
+0: {'loss': 0.8365, 'grad_norm': 0.8285999086461108, 'learning_rate': 2e-05, 'epoch': 0.24}
+0:                                                         

 24%|██▍       | 3510/14725 [1:49:30<5:36:28,  1.80s/it]
 24%|██▍       | 3511/14725 [1:49:32<5:37:11,  1.80s/it]
 24%|██▍       | 3512/14725 [1:49:34<5:37:55,  1.81s/it]
 24%|██▍       | 3513/14725 [1:49:36<5:35:12,  1.79s/it]
 24%|██▍       | 3514/14725 [1:49:38<5:37:25,  1.81s/it]
 24%|██▍       | 3515/14725 [1:49:39<5:35:53,  1.80s/it]
 24%|██▍       | 3516/14725 [1:49:41<5:40:23,  1.82s/it]
 24%|██▍       | 3517/14725 [1:49:43<5:38:43,  1.81s/it]
 24%|██▍       | 3518/14725 [1:49:45<5:36:04,  1.80s/it]
 24%|██▍       | 3519/14725 [1:49:47<5:35:25,  1.80s/it]
 24%|██▍       | 3520/14725 [1:49:49<5:37:56,  1.81s/it]
                                                        

 24%|██▍       | 3520/14725 [1:49:49<5:37:56,  1.81s/it]
 24%|██▍       | 3521/14725 [1:49:50<5:37:25,  1.81s/it]
 24%|██▍       | 3522/14725 [1:49:52<5:38:00,  1.81s/it]
 24%|██▍       | 352
+0: {'loss': 0.8693, 'grad_norm': 0.7933352607031566, 'learning_rate': 2e-05, 'epoch': 0.24}
+0: 3/14725 [1:49:54<5:39:55,  1.82s/it]
 24%|██▍       | 3524/14725 [1:49:56<5:38:31,  1.81s/it]
 24%|██▍       | 3525/14725 [1:49:58<5:41:59,  1.83s/it]
 24%|██▍       | 3526/14725 [1:50:00<5:49:05,  1.87s/it]
 24%|██▍       | 3527/14725 [1:50:01<5:47:05,  1.86s/it]
 24%|██▍       | 3528/14725 [1:50:03<5:45:36,  1.85s/it]
 24%|██▍       | 3529/14725 [1:50:05<5:47:28,  1.86s/it]
 24%|██▍       | 3530/14725 [1:50:07<5:45:29,  1.85s/it]
                                                        

 24%|██▍       | 3530/14725 [1:50:07<5:45:29,  1.85s/it]
 24%|██▍       | 3531/14725 [1:50:09<5:44:39,  1.85s/it]
 24%|██▍       | 3532/14725 [1:50:11<5:43:32,  1.84s/it]
 24%|██▍       | 3533/14725 [1:50:12<5:39:59,  1.82s/it]
 24%|██▍       | 3534/14725 [1:50:14<5:41:12,  1.83s/it]
 24%|██▍       | 3535/14725 [1:50:16<5:41:00,  1.83s/it]
 24%|██▍       | 3536/14725 [1:50:18<5:42:37,  1.84s/it]
 24%|██▍       | 3537/14725 [1:50:20<5:38
+0: {'loss': 0.8564, 'grad_norm': 0.7449568142297627, 'learning_rate': 2e-05, 'epoch': 0.24}
+0: {'loss': 0.8466, 'grad_norm': 0.8220930705796061, 'learning_rate': 2e-05, 'epoch': 0.24}
+0: :04,  1.81s/it]
 24%|██▍       | 3538/14725 [1:50:21<5:33:58,  1.79s/it]
 24%|██▍       | 3539/14725 [1:50:23<5:36:03,  1.80s/it]
 24%|██▍       | 3540/14725 [1:50:25<5:34:49,  1.80s/it]
                                                        

 24%|██▍       | 3540/14725 [1:50:25<5:34:49,  1.80s/it]
 24%|██▍       | 3541/14725 [1:50:27<5:43:44,  1.84s/it]
 24%|██▍       | 3542/14725 [1:50:29<5:41:17,  1.83s/it]
 24%|██▍       | 3543/14725 [1:50:31<5:37:36,  1.81s/it]
 24%|██▍       | 3544/14725 [1:50:32<5:35:43,  1.80s/it]
 24%|██▍       | 3545/14725 [1:50:34<5:37:51,  1.81s/it]
 24%|██▍       | 3546/14725 [1:50:36<5:38:02,  1.81s/it]
 24%|██▍       | 3547/14725 [1:50:38<5:35:55,  1.80s/it]
 24%|██▍       | 3548/14725 [1:50:40<5:36:35,  1.81s/it]
 24%|██▍       | 3549/14725 [1:50:41<5:34:40,  1.80s/it]
 24%|██▍       | 3550/14725 [1:50:43<5:35:48,  1.80s/it]
                                                        

 24%|█�
+0: {'loss': 0.8492, 'grad_norm': 0.8054208251676659, 'learning_rate': 2e-05, 'epoch': 0.24}
+0: �▍       | 3550/14725 [1:50:43<5:35:48,  1.80s/it]
 24%|██▍       | 3551/14725 [1:50:45<5:38:25,  1.82s/it]
 24%|██▍       | 3552/14725 [1:50:47<5:35:37,  1.80s/it]
 24%|██▍       | 3553/14725 [1:50:49<5:33:39,  1.79s/it]
 24%|██▍       | 3554/14725 [1:50:50<5:32:01,  1.78s/it]
 24%|██▍       | 3555/14725 [1:50:52<5:32:28,  1.79s/it]
 24%|██▍       | 3556/14725 [1:50:54<5:35:01,  1.80s/it]
 24%|██▍       | 3557/14725 [1:50:56<5:32:58,  1.79s/it]
 24%|██▍       | 3558/14725 [1:50:58<5:36:20,  1.81s/it]
 24%|██▍       | 3559/14725 [1:50:59<5:36:39,  1.81s/it]
 24%|██▍       | 3560/14725 [1:51:01<5:35:47,  1.80s/it]
                                                        

 24%|██▍       | 3560/14725 [1:51:01<5:35:47,  1.80s/it]
 24%|██▍       | 3561/14725 [1:51:03<5:40:54,  1.83s/it]
 24%|██▍       | 3562/14725 [1:51:05<5:37:54,  1.82s/it]
 24%|██▍       | 3563/14725 [1:51:07<5:39:38,  1.83s/it]
 24%|██▍       | 3564/147
+0: {'loss': 0.8453, 'grad_norm': 0.7691002104982564, 'learning_rate': 2e-05, 'epoch': 0.24}
+0: 25 [1:51:09<5:36:36,  1.81s/it]
 24%|██▍       | 3565/14725 [1:51:10<5:35:57,  1.81s/it]
 24%|██▍       | 3566/14725 [1:51:12<5:36:06,  1.81s/it]
 24%|██▍       | 3567/14725 [1:51:14<5:35:21,  1.80s/it]
 24%|██▍       | 3568/14725 [1:51:16<5:36:33,  1.81s/it]
 24%|██▍       | 3569/14725 [1:51:18<5:34:53,  1.80s/it]
 24%|██▍       | 3570/14725 [1:51:19<5:31:51,  1.78s/it]
                                                        

 24%|██▍       | 3570/14725 [1:51:19<5:31:51,  1.78s/it]
 24%|██▍       | 3571/14725 [1:51:21<5:35:00,  1.80s/it]
 24%|██▍       | 3572/14725 [1:51:23<5:34:53,  1.80s/it]
 24%|██▍       | 3573/14725 [1:51:25<5:36:17,  1.81s/it]
 24%|██▍       | 3574/14725 [1:51:27<5:34:21,  1.80s/it]
 24%|██▍       | 3575/14725 [1:51:28<5:37:18,  1.82s/it]
 24%|██▍       | 3576/14725 [1:51:30<5:36:34,  1.81s/it]
 24%|██▍       | 3577/14725 [1:51:32<5:35:29,  1.81s/it]
 24%|██▍       | 3578/14725 [1:51:34<5:32:05, 
+0: {'loss': 0.8632, 'grad_norm': 0.793096876721054, 'learning_rate': 2e-05, 'epoch': 0.24}
+0: {'loss': 0.8646, 'grad_norm': 0.7243117902654568, 'learning_rate': 2e-05, 'epoch': 0.24}
+0:  1.79s/it]
 24%|██▍       | 3579/14725 [1:51:36<5:33:12,  1.79s/it]
 24%|██▍       | 3580/14725 [1:51:37<5:31:50,  1.79s/it]
                                                        

 24%|██▍       | 3580/14725 [1:51:37<5:31:50,  1.79s/it]
 24%|██▍       | 3581/14725 [1:51:39<5:36:42,  1.81s/it]
 24%|██▍       | 3582/14725 [1:51:41<5:39:44,  1.83s/it]
 24%|██▍       | 3583/14725 [1:51:43<5:37:13,  1.82s/it]
 24%|██▍       | 3584/14725 [1:51:45<5:34:28,  1.80s/it]
 24%|██▍       | 3585/14725 [1:51:46<5:33:29,  1.80s/it]
 24%|██▍       | 3586/14725 [1:51:48<5:40:12,  1.83s/it]
 24%|██▍       | 3587/14725 [1:51:50<5:37:12,  1.82s/it]
 24%|██▍       | 3588/14725 [1:51:52<5:34:37,  1.80s/it]
 24%|██▍       | 3589/14725 [1:51:54<5:32:35,  1.79s/it]
 24%|██▍       | 3590/14725 [1:51:55<5:33:21,  1.80s/it]
                                                        

 24%|██▍       | 3590/14725 [1:51:55<5:33:21,  1.80s/it]
 24%|██▍ 
+0: {'loss': 0.8529, 'grad_norm': 0.7662352222710416, 'learning_rate': 2e-05, 'epoch': 0.24}
+0:       | 3591/14725 [1:51:57<5:32:20,  1.79s/it]
 24%|██▍       | 3592/14725 [1:51:59<5:32:59,  1.79s/it]
 24%|██▍       | 3593/14725 [1:52:01<5:36:54,  1.82s/it]
 24%|██▍       | 3594/14725 [1:52:03<5:38:59,  1.83s/it]
 24%|██▍       | 3595/14725 [1:52:05<5:39:40,  1.83s/it]
 24%|██▍       | 3596/14725 [1:52:06<5:35:31,  1.81s/it]
 24%|██▍       | 3597/14725 [1:52:08<5:34:33,  1.80s/it]
 24%|██▍       | 3598/14725 [1:52:10<5:36:22,  1.81s/it]
 24%|██▍       | 3599/14725 [1:52:12<5:33:12,  1.80s/it]
 24%|██▍       | 3600/14725 [1:52:13<5:31:15,  1.79s/it]
                                                        

 24%|██▍       | 3600/14725 [1:52:13<5:31:15,  1.79s/it]
 24%|██▍       | 3601/14725 [1:52:15<5:36:29,  1.81s/it]
 24%|██▍       | 3602/14725 [1:52:17<5:37:37,  1.82s/it]
 24%|██▍       | 3603/14725 [1:52:19<5:37:33,  1.82s/it]
 24%|██▍       | 3604/14725 [1:52:21<5:34:52,  1.81s/it]
 24%|██▍       | 3605/14725 [1
+0: {'loss': 0.8701, 'grad_norm': 0.8143837024075505, 'learning_rate': 2e-05, 'epoch': 0.25}
+0: :52:23<5:34:42,  1.81s/it]
 24%|██▍       | 3606/14725 [1:52:24<5:37:30,  1.82s/it]
 24%|██▍       | 3607/14725 [1:52:26<5:39:51,  1.83s/it]
 25%|██▍       | 3608/14725 [1:52:28<5:41:10,  1.84s/it]
 25%|██▍       | 3609/14725 [1:52:30<5:38:50,  1.83s/it]
 25%|██▍       | 3610/14725 [1:52:32<5:37:34,  1.82s/it]
                                                        

 25%|██▍       | 3610/14725 [1:52:32<5:37:34,  1.82s/it]
 25%|██▍       | 3611/14725 [1:52:34<5:38:34,  1.83s/it]
 25%|██▍       | 3612/14725 [1:52:35<5:37:18,  1.82s/it]
 25%|██▍       | 3613/14725 [1:52:37<5:38:17,  1.83s/it]
 25%|██▍       | 3614/14725 [1:52:39<5:37:01,  1.82s/it]
 25%|██▍       | 3615/14725 [1:52:41<5:42:43,  1.85s/it]
 25%|██▍       | 3616/14725 [1:52:43<5:44:06,  1.86s/it]
 25%|██▍       | 3617/14725 [1:52:45<5:38:21,  1.83s/it]
 25%|██▍       | 3618/14725 [1:52:46<5:35:59,  1.82s/it]
 25%|██▍       | 3619/14725 [1:52:48<5:43:05,  1.85
+0: {'loss': 0.8275, 'grad_norm': 0.8879752450224702, 'learning_rate': 2e-05, 'epoch': 0.25}
+0: {'loss': 0.8679, 'grad_norm': 0.8166258813959492, 'learning_rate': 2e-05, 'epoch': 0.25}
+0: s/it]
 25%|██▍       | 3620/14725 [1:52:50<5:39:11,  1.83s/it]
                                                        

 25%|██▍       | 3620/14725 [1:52:50<5:39:11,  1.83s/it]
 25%|██▍       | 3621/14725 [1:52:52<5:40:32,  1.84s/it]
 25%|██▍       | 3622/14725 [1:52:54<5:39:34,  1.84s/it]
 25%|██▍       | 3623/14725 [1:52:56<5:37:38,  1.82s/it]
 25%|██▍       | 3624/14725 [1:52:57<5:33:03,  1.80s/it]
 25%|██▍       | 3625/14725 [1:52:59<5:29:51,  1.78s/it]
 25%|██▍       | 3626/14725 [1:53:01<5:32:01,  1.79s/it]
 25%|██▍       | 3627/14725 [1:53:03<5:30:38,  1.79s/it]
 25%|██▍       | 3628/14725 [1:53:04<5:31:11,  1.79s/it]
 25%|██▍       | 3629/14725 [1:53:06<5:36:13,  1.82s/it]
 25%|██▍       | 3630/14725 [1:53:08<5:34:08,  1.81s/it]
                                                        

 25%|██▍       | 3630/14725 [1:53:08<5:34:08,  1.81s/it]
 25%|██▍       | 3631/14725 [1:53:10<5:34:26,  1.81s/it]
 25%|██▍      
+0: {'loss': 0.875, 'grad_norm': 0.792146701160386, 'learning_rate': 2e-05, 'epoch': 0.25}
+0:  | 3632/14725 [1:53:12<5:36:23,  1.82s/it]
 25%|██▍       | 3633/14725 [1:53:14<5:34:07,  1.81s/it]
 25%|██▍       | 3634/14725 [1:53:15<5:33:44,  1.81s/it]
 25%|██▍       | 3635/14725 [1:53:17<5:33:41,  1.81s/it]
 25%|██▍       | 3636/14725 [1:53:19<5:34:11,  1.81s/it]
 25%|██▍       | 3637/14725 [1:53:21<5:36:07,  1.82s/it]
 25%|██▍       | 3638/14725 [1:53:23<5:39:16,  1.84s/it]
 25%|██▍       | 3639/14725 [1:53:25<5:39:35,  1.84s/it]
 25%|██▍       | 3640/14725 [1:53:26<5:38:23,  1.83s/it]
                                                        

 25%|██▍       | 3640/14725 [1:53:26<5:38:23,  1.83s/it]
 25%|██▍       | 3641/14725 [1:53:28<5:37:27,  1.83s/it]
 25%|██▍       | 3642/14725 [1:53:30<5:36:28,  1.82s/it]
 25%|██▍       | 3643/14725 [1:53:32<5:32:55,  1.80s/it]
 25%|██▍       | 3644/14725 [1:53:34<5:32:42,  1.80s/it]
 25%|██▍       | 3645/14725 [1:53:35<5:31:33,  1.80s/it]
 25%|██▍       | 3646/14725 [1:53:3
+0: {'loss': 0.8581, 'grad_norm': 0.7474388967567381, 'learning_rate': 2e-05, 'epoch': 0.25}
+0: 7<5:28:27,  1.78s/it]
 25%|██▍       | 3647/14725 [1:53:39<5:27:19,  1.77s/it]
 25%|██▍       | 3648/14725 [1:53:41<5:28:43,  1.78s/it]
 25%|██▍       | 3649/14725 [1:53:42<5:28:09,  1.78s/it]
 25%|██▍       | 3650/14725 [1:53:44<5:33:13,  1.81s/it]
                                                        

 25%|██▍       | 3650/14725 [1:53:44<5:33:13,  1.81s/it]
 25%|██▍       | 3651/14725 [1:53:46<5:34:19,  1.81s/it]
 25%|██▍       | 3652/14725 [1:53:48<5:33:51,  1.81s/it]
 25%|██▍       | 3653/14725 [1:53:50<5:38:49,  1.84s/it]
 25%|██▍       | 3654/14725 [1:53:52<5:36:13,  1.82s/it]
 25%|██▍       | 3655/14725 [1:53:53<5:36:57,  1.83s/it]
 25%|██▍       | 3656/14725 [1:53:55<5:33:09,  1.81s/it]
 25%|██▍       | 3657/14725 [1:53:57<5:34:59,  1.82s/it]
 25%|██▍       | 3658/14725 [1:53:59<5:42:23,  1.86s/it]
 25%|██▍       | 3659/14725 [1:54:01<5:41:09,  1.85s/it]
 25%|██▍       | 3660/14725 [1:54:03<5:37:25,  1.83s/it]
+0: {'loss': 0.8507, 'grad_norm': 0.7573830992196379, 'learning_rate': 2e-05, 'epoch': 0.25}
+0: {'loss': 0.8473, 'grad_norm': 0.7738595945794212, 'learning_rate': 2e-05, 'epoch': 0.25}
+0: 
                                                        

 25%|██▍       | 3660/14725 [1:54:03<5:37:25,  1.83s/it]
 25%|██▍       | 3661/14725 [1:54:04<5:33:03,  1.81s/it]
 25%|██▍       | 3662/14725 [1:54:06<5:30:04,  1.79s/it]
 25%|██▍       | 3663/14725 [1:54:08<5:30:00,  1.79s/it]
 25%|██▍       | 3664/14725 [1:54:10<5:31:11,  1.80s/it]
 25%|██▍       | 3665/14725 [1:54:12<5:35:13,  1.82s/it]
 25%|██▍       | 3666/14725 [1:54:13<5:34:44,  1.82s/it]
 25%|██▍       | 3667/14725 [1:54:15<5:32:05,  1.80s/it]
 25%|██▍       | 3668/14725 [1:54:17<5:33:42,  1.81s/it]
 25%|██▍       | 3669/14725 [1:54:19<5:36:48,  1.83s/it]
 25%|██▍       | 3670/14725 [1:54:21<5:36:05,  1.82s/it]
                                                        

 25%|██▍       | 3670/14725 [1:54:21<5:36:05,  1.82s/it]
 25%|██▍       | 3671/14725 [1:54:23<5:39:09,  1.84s/it]
 25%|██▍       | 3672/14725 [1:54:24<5:44:20,  1.87s/it]
 25%|██▍       | 36
+0: {'loss': 0.8656, 'grad_norm': 0.7210623289273229, 'learning_rate': 2e-05, 'epoch': 0.25}
+0: 73/14725 [1:54:26<5:44:48,  1.87s/it]
 25%|██▍       | 3674/14725 [1:54:28<5:45:45,  1.88s/it]
 25%|██▍       | 3675/14725 [1:54:30<5:47:47,  1.89s/it]
 25%|██▍       | 3676/14725 [1:54:32<5:39:59,  1.85s/it]
 25%|██▍       | 3677/14725 [1:54:34<5:37:36,  1.83s/it]
 25%|██▍       | 3678/14725 [1:54:36<5:35:13,  1.82s/it]
 25%|██▍       | 3679/14725 [1:54:37<5:36:55,  1.83s/it]
 25%|██▍       | 3680/14725 [1:54:39<5:37:30,  1.83s/it]
                                                        

 25%|██▍       | 3680/14725 [1:54:39<5:37:30,  1.83s/it]
 25%|██▍       | 3681/14725 [1:54:41<5:35:26,  1.82s/it]
 25%|██▌       | 3682/14725 [1:54:43<5:33:33,  1.81s/it]
 25%|██▌       | 3683/14725 [1:54:45<5:29:48,  1.79s/it]
 25%|██▌       | 3684/14725 [1:54:46<5:29:31,  1.79s/it]
 25%|██▌       | 3685/14725 [1:54:48<5:29:33,  1.79s/it]
 25%|██▌       | 3686/14725 [1:54:50<5:28:37,  1.79s/it]
 25%|██▌       | 3687/14725 [1:54:52<5:2
+0: {'loss': 0.8337, 'grad_norm': 0.8585983905320876, 'learning_rate': 2e-05, 'epoch': 0.25}
+0: {'loss': 0.8423, 'grad_norm': 0.7940055949348231, 'learning_rate': 2e-05, 'epoch': 0.25}
+0: 8:48,  1.79s/it]
 25%|██▌       | 3688/14725 [1:54:53<5:26:59,  1.78s/it]
 25%|██▌       | 3689/14725 [1:54:55<5:26:55,  1.78s/it]
 25%|██▌       | 3690/14725 [1:54:57<5:28:12,  1.78s/it]
                                                        

 25%|██▌       | 3690/14725 [1:54:57<5:28:12,  1.78s/it]
 25%|██▌       | 3691/14725 [1:54:59<5:26:42,  1.78s/it]
 25%|██▌       | 3692/14725 [1:55:01<5:30:38,  1.80s/it]
 25%|██▌       | 3693/14725 [1:55:02<5:29:11,  1.79s/it]
 25%|██▌       | 3694/14725 [1:55:04<5:34:37,  1.82s/it]
 25%|██▌       | 3695/14725 [1:55:06<5:32:12,  1.81s/it]
 25%|██▌       | 3696/14725 [1:55:08<5:34:37,  1.82s/it]
 25%|██▌       | 3697/14725 [1:55:10<5:37:42,  1.84s/it]
 25%|██▌       | 3698/14725 [1:55:12<5:37:08,  1.83s/it]
 25%|██▌       | 3699/14725 [1:55:13<5:33:01,  1.81s/it]
 25%|██▌       | 3700/14725 [1:55:15<5:31:32,  1.80s/it]
                                                        

 25%|█�
+0: {'loss': 0.8346, 'grad_norm': 0.7396328298352755, 'learning_rate': 2e-05, 'epoch': 0.25}
+0: ��▌       | 3700/14725 [1:55:15<5:31:32,  1.80s/it]
 25%|██▌       | 3701/14725 [1:55:17<5:31:34,  1.80s/it]
 25%|██▌       | 3702/14725 [1:55:19<5:33:35,  1.82s/it]
 25%|██▌       | 3703/14725 [1:55:21<5:29:43,  1.79s/it]
 25%|██▌       | 3704/14725 [1:55:22<5:30:01,  1.80s/it]
 25%|██▌       | 3705/14725 [1:55:24<5:32:09,  1.81s/it]
 25%|██▌       | 3706/14725 [1:55:26<5:29:43,  1.80s/it]
 25%|██▌       | 3707/14725 [1:55:28<5:28:44,  1.79s/it]
 25%|██▌       | 3708/14725 [1:55:29<5:27:40,  1.78s/it]
 25%|██▌       | 3709/14725 [1:55:31<5:31:44,  1.81s/it]
 25%|██▌       | 3710/14725 [1:55:33<5:32:14,  1.81s/it]
                                                        

 25%|██▌       | 3710/14725 [1:55:33<5:32:14,  1.81s/it]
 25%|██▌       | 3711/14725 [1:55:35<5:31:30,  1.81s/it]
 25%|██▌       | 3712/14725 [1:55:37<5:32:06,  1.81s/it]
 25%|██▌       | 3713/14725 [1:55:39<5:36:56,  1.84s/it]
 25%|██▌       | 3714/14
+0: {'loss': 0.8493, 'grad_norm': 0.8102154282753686, 'learning_rate': 2e-05, 'epoch': 0.25}
+0: 725 [1:55:41<5:35:41,  1.83s/it]
 25%|██▌       | 3715/14725 [1:55:42<5:35:33,  1.83s/it]
 25%|██▌       | 3716/14725 [1:55:44<5:33:39,  1.82s/it]
 25%|██▌       | 3717/14725 [1:55:46<5:29:48,  1.80s/it]
 25%|██▌       | 3718/14725 [1:55:48<5:28:14,  1.79s/it]
 25%|██▌       | 3719/14725 [1:55:49<5:25:50,  1.78s/it]
 25%|██▌       | 3720/14725 [1:55:51<5:31:12,  1.81s/it]
                                                        

 25%|██▌       | 3720/14725 [1:55:51<5:31:12,  1.81s/it]
 25%|██▌       | 3721/14725 [1:55:53<5:28:46,  1.79s/it]
 25%|██▌       | 3722/14725 [1:55:55<5:26:24,  1.78s/it]
 25%|██▌       | 3723/14725 [1:55:57<5:27:53,  1.79s/it]
 25%|██▌       | 3724/14725 [1:55:58<5:30:00,  1.80s/it]
 25%|██▌       | 3725/14725 [1:56:00<5:30:09,  1.80s/it]
 25%|██▌       | 3726/14725 [1:56:02<5:32:41,  1.81s/it]
 25%|██▌       | 3727/14725 [1:56:04<5:32:40,  1.81s/it]
 25%|██▌       | 3728/14725 [1:56:06<5:37:22,
+0: {'loss': 0.8417, 'grad_norm': 0.7636086554784745, 'learning_rate': 2e-05, 'epoch': 0.25}
+0: {'loss': 0.8378, 'grad_norm': 0.7331004820550618, 'learning_rate': 2e-05, 'epoch': 0.25}
+0:   1.84s/it]
 25%|██▌       | 3729/14725 [1:56:08<5:32:31,  1.81s/it]
 25%|██▌       | 3730/14725 [1:56:09<5:35:15,  1.83s/it]
                                                        

 25%|██▌       | 3730/14725 [1:56:09<5:35:15,  1.83s/it]
 25%|██▌       | 3731/14725 [1:56:11<5:39:47,  1.85s/it]
 25%|██▌       | 3732/14725 [1:56:13<5:36:19,  1.84s/it]
 25%|██▌       | 3733/14725 [1:56:15<5:32:20,  1.81s/it]
 25%|██▌       | 3734/14725 [1:56:17<5:31:19,  1.81s/it]
 25%|██▌       | 3735/14725 [1:56:18<5:32:47,  1.82s/it]
 25%|██▌       | 3736/14725 [1:56:20<5:34:59,  1.83s/it]
 25%|██▌       | 3737/14725 [1:56:22<5:30:47,  1.81s/it]
 25%|██▌       | 3738/14725 [1:56:24<5:29:31,  1.80s/it]
 25%|██▌       | 3739/14725 [1:56:26<5:27:19,  1.79s/it]
 25%|██▌       | 3740/14725 [1:56:27<5:29:22,  1.80s/it]
                                                        

 25%|██▌       | 3740/14725 [1:56:27<5:29:22,  1.80s/it]
 25%|██▌
+0: {'loss': 0.8557, 'grad_norm': 0.7817822374847302, 'learning_rate': 2e-05, 'epoch': 0.25}
+0:        | 3741/14725 [1:56:29<5:27:17,  1.79s/it]
 25%|██▌       | 3742/14725 [1:56:31<5:32:36,  1.82s/it]
 25%|██▌       | 3743/14725 [1:56:33<5:28:55,  1.80s/it]
 25%|██▌       | 3744/14725 [1:56:35<5:29:51,  1.80s/it]
 25%|██▌       | 3745/14725 [1:56:36<5:27:28,  1.79s/it]
 25%|██▌       | 3746/14725 [1:56:38<5:27:42,  1.79s/it]
 25%|██▌       | 3747/14725 [1:56:40<5:27:44,  1.79s/it]
 25%|██▌       | 3748/14725 [1:56:42<5:32:33,  1.82s/it]
 25%|██▌       | 3749/14725 [1:56:44<5:33:00,  1.82s/it]
 25%|██▌       | 3750/14725 [1:56:46<5:36:36,  1.84s/it]
                                                        

 25%|██▌       | 3750/14725 [1:56:46<5:36:36,  1.84s/it]
 25%|██▌       | 3751/14725 [1:56:47<5:33:45,  1.82s/it]
 25%|██▌       | 3752/14725 [1:56:49<5:33:46,  1.83s/it]
 25%|██▌       | 3753/14725 [1:56:51<5:32:01,  1.82s/it]
 25%|██▌       | 3754/14725 [1:56:53<5:30:30,  1.81s/it]
 26%|██▌       | 3755/14725 [
+0: {'loss': 0.858, 'grad_norm': 0.842272983962661, 'learning_rate': 2e-05, 'epoch': 0.26}
+0: 1:56:55<5:28:33,  1.80s/it]
 26%|██▌       | 3756/14725 [1:56:56<5:27:26,  1.79s/it]
 26%|██▌       | 3757/14725 [1:56:58<5:31:33,  1.81s/it]
 26%|██▌       | 3758/14725 [1:57:00<5:31:55,  1.82s/it]
 26%|██▌       | 3759/14725 [1:57:02<5:32:32,  1.82s/it]
 26%|██▌       | 3760/14725 [1:57:04<5:37:37,  1.85s/it]
                                                        

 26%|██▌       | 3760/14725 [1:57:04<5:37:37,  1.85s/it]
 26%|██▌       | 3761/14725 [1:57:06<5:34:29,  1.83s/it]
 26%|██▌       | 3762/14725 [1:57:07<5:35:28,  1.84s/it]
 26%|██▌       | 3763/14725 [1:57:09<5:31:00,  1.81s/it]
 26%|██▌       | 3764/14725 [1:57:11<5:27:25,  1.79s/it]
 26%|██▌       | 3765/14725 [1:57:13<5:27:05,  1.79s/it]
 26%|██▌       | 3766/14725 [1:57:15<5:27:31,  1.79s/it]
 26%|██▌       | 3767/14725 [1:57:16<5:31:54,  1.82s/it]
 26%|██▌       | 3768/14725 [1:57:18<5:33:27,  1.83s/it]
 26%|██▌       | 3769/14725 [1:57:20<5:31:34,  1.8
+0: {'loss': 0.8581, 'grad_norm': 0.722334440819762, 'learning_rate': 2e-05, 'epoch': 0.26}
+0: {'loss': 0.8742, 'grad_norm': 0.7793654515477567, 'learning_rate': 2e-05, 'epoch': 0.26}
+0: 2s/it]
 26%|██▌       | 3770/14725 [1:57:22<5:37:07,  1.85s/it]
                                                        

 26%|██▌       | 3770/14725 [1:57:22<5:37:07,  1.85s/it]
 26%|██▌       | 3771/14725 [1:57:24<5:34:53,  1.83s/it]
 26%|██▌       | 3772/14725 [1:57:26<5:33:34,  1.83s/it]
 26%|██▌       | 3773/14725 [1:57:27<5:33:36,  1.83s/it]
 26%|██▌       | 3774/14725 [1:57:29<5:34:55,  1.84s/it]
 26%|██▌       | 3775/14725 [1:57:31<5:34:06,  1.83s/it]
 26%|██▌       | 3776/14725 [1:57:33<5:31:05,  1.81s/it]
 26%|██▌       | 3777/14725 [1:57:35<5:36:01,  1.84s/it]
 26%|██▌       | 3778/14725 [1:57:37<5:38:27,  1.86s/it]
 26%|██▌       | 3779/14725 [1:57:38<5:33:38,  1.83s/it]
 26%|██▌       | 3780/14725 [1:57:40<5:33:27,  1.83s/it]
                                                        

 26%|██▌       | 3780/14725 [1:57:40<5:33:27,  1.83s/it]
 26%|██▌       | 3781/14725 [1:57:42<5:32:17,  1.82s/it]
 26%|██▌     
+0: {'loss': 0.8602, 'grad_norm': 0.8284166767750774, 'learning_rate': 2e-05, 'epoch': 0.26}
+0:   | 3782/14725 [1:57:44<5:29:32,  1.81s/it]
 26%|██▌       | 3783/14725 [1:57:46<5:29:57,  1.81s/it]
 26%|██▌       | 3784/14725 [1:57:47<5:28:02,  1.80s/it]
 26%|██▌       | 3785/14725 [1:57:49<5:28:42,  1.80s/it]
 26%|██▌       | 3786/14725 [1:57:51<5:31:18,  1.82s/it]
 26%|██▌       | 3787/14725 [1:57:53<5:28:29,  1.80s/it]
 26%|██▌       | 3788/14725 [1:57:55<5:27:07,  1.79s/it]
 26%|██▌       | 3789/14725 [1:57:56<5:29:48,  1.81s/it]
 26%|██▌       | 3790/14725 [1:57:58<5:30:07,  1.81s/it]
                                                        

 26%|██▌       | 3790/14725 [1:57:58<5:30:07,  1.81s/it]
 26%|██▌       | 3791/14725 [1:58:00<5:31:11,  1.82s/it]
 26%|██▌       | 3792/14725 [1:58:02<5:27:12,  1.80s/it]
 26%|██▌       | 3793/14725 [1:58:04<5:23:54,  1.78s/it]
 26%|██▌       | 3794/14725 [1:58:05<5:25:38,  1.79s/it]
 26%|██▌       | 3795/14725 [1:58:07<5:27:50,  1.80s/it]
 26%|██▌       | 3796/14725 [1:58:
+0: {'loss': 0.8661, 'grad_norm': 0.7761342975397119, 'learning_rate': 2e-05, 'epoch': 0.26}
+0: 09<5:26:05,  1.79s/it]
 26%|██▌       | 3797/14725 [1:58:11<5:25:42,  1.79s/it]
 26%|██▌       | 3798/14725 [1:58:13<5:24:55,  1.78s/it]
 26%|██▌       | 3799/14725 [1:58:14<5:22:31,  1.77s/it]
 26%|██▌       | 3800/14725 [1:58:16<5:22:59,  1.77s/it]
                                                        

 26%|██▌       | 3800/14725 [1:58:16<5:22:59,  1.77s/it]
 26%|██▌       | 3801/14725 [1:58:18<5:27:42,  1.80s/it]
 26%|██▌       | 3802/14725 [1:58:20<5:28:30,  1.80s/it]
 26%|██▌       | 3803/14725 [1:58:22<5:26:27,  1.79s/it]
 26%|██▌       | 3804/14725 [1:58:23<5:28:41,  1.81s/it]
 26%|██▌       | 3805/14725 [1:58:25<5:26:30,  1.79s/it]
 26%|██▌       | 3806/14725 [1:58:27<5:29:23,  1.81s/it]
 26%|██▌       | 3807/14725 [1:58:29<5:28:07,  1.80s/it]
 26%|██▌       | 3808/14725 [1:58:31<5:27:07,  1.80s/it]
 26%|██▌       | 3809/14725 [1:58:32<5:24:19,  1.78s/it]
 26%|██▌       | 3810/14725 [1:58:34<5:27:43,  1.80s/it
+0: {'loss': 0.8389, 'grad_norm': 0.7633514094872496, 'learning_rate': 2e-05, 'epoch': 0.26}
+0: {'loss': 0.8573, 'grad_norm': 0.7849104840936991, 'learning_rate': 2e-05, 'epoch': 0.26}
+0: ]
                                                        

 26%|██▌       | 3810/14725 [1:58:34<5:27:43,  1.80s/it]
 26%|██▌       | 3811/14725 [1:58:36<5:31:49,  1.82s/it]
 26%|██▌       | 3812/14725 [1:58:38<5:28:49,  1.81s/it]
 26%|██▌       | 3813/14725 [1:58:40<5:28:53,  1.81s/it]
 26%|██▌       | 3814/14725 [1:58:41<5:29:39,  1.81s/it]
 26%|██▌       | 3815/14725 [1:58:43<5:30:56,  1.82s/it]
 26%|██▌       | 3816/14725 [1:58:45<5:30:41,  1.82s/it]
 26%|██▌       | 3817/14725 [1:58:47<5:31:53,  1.83s/it]
 26%|██▌       | 3818/14725 [1:58:49<5:32:51,  1.83s/it]
 26%|██▌       | 3819/14725 [1:58:51<5:35:17,  1.84s/it]
 26%|██▌       | 3820/14725 [1:58:52<5:32:25,  1.83s/it]
                                                        

 26%|██▌       | 3820/14725 [1:58:52<5:32:25,  1.83s/it]
 26%|██▌       | 3821/14725 [1:58:54<5:32:34,  1.83s/it]
 26%|██▌       | 3822/14725 [1:58:56<5:34:08,  1.84s/it]
 26%|██▌       | 3
+0: {'loss': 0.8577, 'grad_norm': 0.7543125765934673, 'learning_rate': 2e-05, 'epoch': 0.26}
+0: 823/14725 [1:58:58<5:34:41,  1.84s/it]
 26%|██▌       | 3824/14725 [1:59:00<5:32:12,  1.83s/it]
 26%|██▌       | 3825/14725 [1:59:02<5:35:28,  1.85s/it]
 26%|██▌       | 3826/14725 [1:59:03<5:32:26,  1.83s/it]
 26%|██▌       | 3827/14725 [1:59:05<5:34:01,  1.84s/it]
 26%|██▌       | 3828/14725 [1:59:07<5:35:51,  1.85s/it]
 26%|██▌       | 3829/14725 [1:59:09<5:36:07,  1.85s/it]
 26%|██▌       | 3830/14725 [1:59:11<5:33:22,  1.84s/it]
                                                        

 26%|██▌       | 3830/14725 [1:59:11<5:33:22,  1.84s/it]
 26%|██▌       | 3831/14725 [1:59:13<5:33:00,  1.83s/it]
 26%|██▌       | 3832/14725 [1:59:14<5:29:55,  1.82s/it]
 26%|██▌       | 3833/14725 [1:59:16<5:30:32,  1.82s/it]
 26%|██▌       | 3834/14725 [1:59:18<5:27:45,  1.81s/it]
 26%|██▌       | 3835/14725 [1:59:20<5:26:35,  1.80s/it]
 26%|██▌       | 3836/14725 [1:59:22<5:29:02,  1.81s/it]
 26%|██▌       | 3837/14725 [1:59:23<5:
+0: {'loss': 0.8518, 'grad_norm': 0.797638847145999, 'learning_rate': 2e-05, 'epoch': 0.26}
+0: {'loss': 0.8499, 'grad_norm': 0.7372098844290835, 'learning_rate': 2e-05, 'epoch': 0.26}
+0: 27:01,  1.80s/it]
 26%|██▌       | 3838/14725 [1:59:25<5:27:43,  1.81s/it]
 26%|██▌       | 3839/14725 [1:59:27<5:25:38,  1.79s/it]
 26%|██▌       | 3840/14725 [1:59:29<5:29:00,  1.81s/it]
                                                        

 26%|██▌       | 3840/14725 [1:59:29<5:29:00,  1.81s/it]
 26%|██▌       | 3841/14725 [1:59:31<5:29:06,  1.81s/it]
 26%|██▌       | 3842/14725 [1:59:33<5:31:17,  1.83s/it]
 26%|██▌       | 3843/14725 [1:59:34<5:31:03,  1.83s/it]
 26%|██▌       | 3844/14725 [1:59:36<5:47:19,  1.92s/it]
 26%|██▌       | 3845/14725 [1:59:39<5:54:43,  1.96s/it]
 26%|██▌       | 3846/14725 [1:59:40<5:47:00,  1.91s/it]
 26%|██▌       | 3847/14725 [1:59:42<5:38:22,  1.87s/it]
 26%|██▌       | 3848/14725 [1:59:44<5:36:32,  1.86s/it]
 26%|██▌       | 3849/14725 [1:59:46<5:32:13,  1.83s/it]
 26%|██▌       | 3850/14725 [1:59:47<5:28:23,  1.81s/it]
                                                        

 26%|█
+0: {'loss': 0.8383, 'grad_norm': 0.7758204547532628, 'learning_rate': 2e-05, 'epoch': 0.26}
+0: █▌       | 3850/14725 [1:59:48<5:28:23,  1.81s/it]
 26%|██▌       | 3851/14725 [1:59:49<5:27:40,  1.81s/it]
 26%|██▌       | 3852/14725 [1:59:51<5:24:54,  1.79s/it]
 26%|██▌       | 3853/14725 [1:59:53<5:23:14,  1.78s/it]
 26%|██▌       | 3854/14725 [1:59:55<5:20:56,  1.77s/it]
 26%|██▌       | 3855/14725 [1:59:56<5:23:41,  1.79s/it]
 26%|██▌       | 3856/14725 [1:59:58<5:22:21,  1.78s/it]
 26%|██▌       | 3857/14725 [2:00:00<5:21:12,  1.77s/it]
 26%|██▌       | 3858/14725 [2:00:02<5:23:12,  1.78s/it]
 26%|██▌       | 3859/14725 [2:00:03<5:22:08,  1.78s/it]
 26%|██▌       | 3860/14725 [2:00:05<5:21:51,  1.78s/it]
                                                        

 26%|██▌       | 3860/14725 [2:00:05<5:21:51,  1.78s/it]
 26%|██▌       | 3861/14725 [2:00:07<5:24:41,  1.79s/it]
 26%|██▌       | 3862/14725 [2:00:09<5:23:29,  1.79s/it]
 26%|██▌       | 3863/14725 [2:00:11<5:24:46,  1.79s/it]
 26%|██▌       | 3864/1
+0: {'loss': 0.8461, 'grad_norm': 0.7609621509102943, 'learning_rate': 2e-05, 'epoch': 0.26}
+0: 4725 [2:00:12<5:23:28,  1.79s/it]
 26%|██▌       | 3865/14725 [2:00:14<5:26:22,  1.80s/it]
 26%|██▋       | 3866/14725 [2:00:16<5:24:36,  1.79s/it]
 26%|██▋       | 3867/14725 [2:00:18<5:26:46,  1.81s/it]
 26%|██▋       | 3868/14725 [2:00:20<5:23:43,  1.79s/it]
 26%|██▋       | 3869/14725 [2:00:21<5:25:17,  1.80s/it]
 26%|██▋       | 3870/14725 [2:00:23<5:25:06,  1.80s/it]
                                                        

 26%|██▋       | 3870/14725 [2:00:23<5:25:06,  1.80s/it]
 26%|██▋       | 3871/14725 [2:00:25<5:26:42,  1.81s/it]
 26%|██▋       | 3872/14725 [2:00:27<5:24:03,  1.79s/it]
 26%|██▋       | 3873/14725 [2:00:29<5:23:35,  1.79s/it]
 26%|██▋       | 3874/14725 [2:00:30<5:23:02,  1.79s/it]
 26%|██▋       | 3875/14725 [2:00:32<5:25:06,  1.80s/it]
 26%|██▋       | 3876/14725 [2:00:34<5:31:21,  1.83s/it]
 26%|██▋       | 3877/14725 [2:00:36<5:31:05,  1.83s/it]
 26%|██▋       | 3878/14725 [2:00:38<5:31:17
+0: {'loss': 0.845, 'grad_norm': 0.7775379078306092, 'learning_rate': 2e-05, 'epoch': 0.26}
+0: {'loss': 0.8496, 'grad_norm': 0.7384510359916601, 'learning_rate': 2e-05, 'epoch': 0.26}
+0: ,  1.83s/it]
 26%|██▋       | 3879/14725 [2:00:40<5:27:09,  1.81s/it]
 26%|██▋       | 3880/14725 [2:00:41<5:31:36,  1.83s/it]
                                                        

 26%|██▋       | 3880/14725 [2:00:41<5:31:36,  1.83s/it]
 26%|██▋       | 3881/14725 [2:00:43<5:32:54,  1.84s/it]
 26%|██▋       | 3882/14725 [2:00:45<5:36:53,  1.86s/it]
 26%|██▋       | 3883/14725 [2:00:47<5:36:24,  1.86s/it]
 26%|██▋       | 3884/14725 [2:00:49<5:34:45,  1.85s/it]
 26%|██▋       | 3885/14725 [2:00:51<5:32:55,  1.84s/it]
 26%|██▋       | 3886/14725 [2:00:53<5:32:28,  1.84s/it]
 26%|██▋       | 3887/14725 [2:00:54<5:28:47,  1.82s/it]
 26%|██▋       | 3888/14725 [2:00:56<5:33:00,  1.84s/it]
 26%|██▋       | 3889/14725 [2:00:58<5:31:22,  1.83s/it]
 26%|██▋       | 3890/14725 [2:01:00<5:29:24,  1.82s/it]
                                                        

 26%|██▋       | 3890/14725 [2:01:00<5:29:24,  1.82s/it]
 26%|██�
+0: {'loss': 0.8552, 'grad_norm': 0.8635143782719947, 'learning_rate': 2e-05, 'epoch': 0.26}
+0: �       | 3891/14725 [2:01:02<5:30:19,  1.83s/it]
 26%|██▋       | 3892/14725 [2:01:03<5:26:59,  1.81s/it]
 26%|██▋       | 3893/14725 [2:01:05<5:27:21,  1.81s/it]
 26%|██▋       | 3894/14725 [2:01:07<5:24:03,  1.80s/it]
 26%|██▋       | 3895/14725 [2:01:09<5:26:51,  1.81s/it]
 26%|██▋       | 3896/14725 [2:01:11<5:29:42,  1.83s/it]
 26%|██▋       | 3897/14725 [2:01:12<5:25:53,  1.81s/it]
 26%|██▋       | 3898/14725 [2:01:14<5:26:19,  1.81s/it]
 26%|██▋       | 3899/14725 [2:01:16<5:26:25,  1.81s/it]
 26%|██▋       | 3900/14725 [2:01:18<5:25:44,  1.81s/it]
                                                        

 26%|██▋       | 3900/14725 [2:01:18<5:25:44,  1.81s/it]
 26%|██▋       | 3901/14725 [2:01:20<5:25:00,  1.80s/it]
 26%|██▋       | 3902/14725 [2:01:22<5:28:10,  1.82s/it]
 27%|██▋       | 3903/14725 [2:01:23<5:24:55,  1.80s/it]
 27%|██▋       | 3904/14725 [2:01:25<5:28:58,  1.82s/it]
 27%|██▋       | 3905/14725 
+0: {'loss': 0.8539, 'grad_norm': 0.8241702320451613, 'learning_rate': 2e-05, 'epoch': 0.27}
+0: [2:01:27<5:29:45,  1.83s/it]
 27%|██▋       | 3906/14725 [2:01:29<5:25:48,  1.81s/it]
 27%|██▋       | 3907/14725 [2:01:31<5:28:08,  1.82s/it]
 27%|██▋       | 3908/14725 [2:01:32<5:28:08,  1.82s/it]
 27%|██▋       | 3909/14725 [2:01:34<5:24:01,  1.80s/it]
 27%|██▋       | 3910/14725 [2:01:36<5:22:39,  1.79s/it]
                                                        

 27%|██▋       | 3910/14725 [2:01:36<5:22:39,  1.79s/it]
 27%|██▋       | 3911/14725 [2:01:38<5:21:30,  1.78s/it]
 27%|██▋       | 3912/14725 [2:01:40<5:21:10,  1.78s/it]
 27%|██▋       | 3913/14725 [2:01:41<5:26:16,  1.81s/it]
 27%|██▋       | 3914/14725 [2:01:43<5:27:14,  1.82s/it]
 27%|██▋       | 3915/14725 [2:01:45<5:25:11,  1.80s/it]
 27%|██▋       | 3916/14725 [2:01:47<5:24:08,  1.80s/it]
 27%|██▋       | 3917/14725 [2:01:49<5:21:46,  1.79s/it]
 27%|██▋       | 3918/14725 [2:01:50<5:23:42,  1.80s/it]
 27%|██▋       | 3919/14725 [2:01:52<5:20:29,  1.
+0: {'loss': 0.8479, 'grad_norm': 0.7526649578309755, 'learning_rate': 2e-05, 'epoch': 0.27}
+0: {'loss': 0.8397, 'grad_norm': 0.8016126240506382, 'learning_rate': 2e-05, 'epoch': 0.27}
+0: 78s/it]
 27%|██▋       | 3920/14725 [2:01:54<5:26:11,  1.81s/it]
                                                        

 27%|██▋       | 3920/14725 [2:01:54<5:26:11,  1.81s/it]
 27%|██▋       | 3921/14725 [2:01:56<5:26:56,  1.82s/it]
 27%|██▋       | 3922/14725 [2:01:58<5:23:44,  1.80s/it]
 27%|██▋       | 3923/14725 [2:01:59<5:24:38,  1.80s/it]
 27%|██▋       | 3924/14725 [2:02:01<5:22:08,  1.79s/it]
 27%|██▋       | 3925/14725 [2:02:03<5:23:04,  1.79s/it]
 27%|██▋       | 3926/14725 [2:02:05<5:21:52,  1.79s/it]
 27%|██▋       | 3927/14725 [2:02:07<5:25:02,  1.81s/it]
 27%|██▋       | 3928/14725 [2:02:08<5:21:50,  1.79s/it]
 27%|██▋       | 3929/14725 [2:02:10<5:20:20,  1.78s/it]
 27%|██▋       | 3930/14725 [2:02:12<5:19:02,  1.77s/it]
                                                        

 27%|██▋       | 3930/14725 [2:02:12<5:19:02,  1.77s/it]
 27%|██▋       | 3931/14725 [2:02:14<5:22:22,  1.79s/it]
 27%|██▋    
+0: {'loss': 0.8466, 'grad_norm': 0.7538037380592735, 'learning_rate': 2e-05, 'epoch': 0.27}
+0:    | 3932/14725 [2:02:15<5:20:42,  1.78s/it]
 27%|██▋       | 3933/14725 [2:02:17<5:21:07,  1.79s/it]
 27%|██▋       | 3934/14725 [2:02:19<5:23:29,  1.80s/it]
 27%|██▋       | 3935/14725 [2:02:21<5:22:13,  1.79s/it]
 27%|██▋       | 3936/14725 [2:02:23<5:22:17,  1.79s/it]
 27%|██▋       | 3937/14725 [2:02:24<5:24:45,  1.81s/it]
 27%|██▋       | 3938/14725 [2:02:26<5:25:34,  1.81s/it]
 27%|██▋       | 3939/14725 [2:02:28<5:25:49,  1.81s/it]
 27%|██▋       | 3940/14725 [2:02:30<5:28:12,  1.83s/it]
                                                        

 27%|██▋       | 3940/14725 [2:02:30<5:28:12,  1.83s/it]
 27%|██▋       | 3941/14725 [2:02:32<5:27:25,  1.82s/it]
 27%|██▋       | 3942/14725 [2:02:34<5:25:41,  1.81s/it]
 27%|██▋       | 3943/14725 [2:02:35<5:22:57,  1.80s/it]
 27%|██▋       | 3944/14725 [2:02:37<5:22:24,  1.79s/it]
 27%|██▋       | 3945/14725 [2:02:39<5:20:12,  1.78s/it]
 27%|██▋       | 3946/14725 [2:02
+0: {'loss': 0.8627, 'grad_norm': 0.7620894238499436, 'learning_rate': 2e-05, 'epoch': 0.27}
+0: :41<5:21:37,  1.79s/it]
 27%|██▋       | 3947/14725 [2:02:42<5:20:57,  1.79s/it]
 27%|██▋       | 3948/14725 [2:02:44<5:22:55,  1.80s/it]
 27%|██▋       | 3949/14725 [2:02:46<5:23:47,  1.80s/it]
 27%|██▋       | 3950/14725 [2:02:48<5:23:05,  1.80s/it]
                                                        

 27%|██▋       | 3950/14725 [2:02:48<5:23:05,  1.80s/it]
 27%|██▋       | 3951/14725 [2:02:50<5:25:51,  1.81s/it]
 27%|██▋       | 3952/14725 [2:02:52<5:26:33,  1.82s/it]
 27%|██▋       | 3953/14725 [2:02:53<5:25:20,  1.81s/it]
 27%|██▋       | 3954/14725 [2:02:55<5:23:06,  1.80s/it]
 27%|██▋       | 3955/14725 [2:02:57<5:22:13,  1.80s/it]
 27%|██▋       | 3956/14725 [2:02:59<5:22:43,  1.80s/it]
 27%|██▋       | 3957/14725 [2:03:01<5:26:20,  1.82s/it]
 27%|██▋       | 3958/14725 [2:03:02<5:24:58,  1.81s/it]
 27%|██▋       | 3959/14725 [2:03:04<5:27:39,  1.83s/it]
 27%|██▋       | 3960/14725 [2:03:06<5:27:00,  1.82s/i
+0: {'loss': 0.8434, 'grad_norm': 0.8404527200715345, 'learning_rate': 2e-05, 'epoch': 0.27}
+0: {'loss': 0.8441, 'grad_norm': 0.7949137862642309, 'learning_rate': 2e-05, 'epoch': 0.27}
+0: t]
                                                        

 27%|██▋       | 3960/14725 [2:03:06<5:27:00,  1.82s/it]
 27%|██▋       | 3961/14725 [2:03:08<5:27:11,  1.82s/it]
 27%|██▋       | 3962/14725 [2:03:10<5:29:22,  1.84s/it]
 27%|██▋       | 3963/14725 [2:03:12<5:27:06,  1.82s/it]
 27%|██▋       | 3964/14725 [2:03:13<5:26:54,  1.82s/it]
 27%|██▋       | 3965/14725 [2:03:15<5:24:09,  1.81s/it]
 27%|██▋       | 3966/14725 [2:03:17<5:26:47,  1.82s/it]
 27%|██▋       | 3967/14725 [2:03:19<5:26:06,  1.82s/it]
 27%|██▋       | 3968/14725 [2:03:21<5:26:29,  1.82s/it]
 27%|██▋       | 3969/14725 [2:03:22<5:25:28,  1.82s/it]
 27%|██▋       | 3970/14725 [2:03:24<5:27:13,  1.83s/it]
                                                        

 27%|██▋       | 3970/14725 [2:03:24<5:27:13,  1.83s/it]
 27%|██▋       | 3971/14725 [2:03:26<5:23:47,  1.81s/it]
 27%|██▋       | 3972/14725 [2:03:28<5:22:04,  1.80s/it]
 27%|██▋       | 
+0: {'loss': 0.8489, 'grad_norm': 0.7877949246066621, 'learning_rate': 2e-05, 'epoch': 0.27}
+0: 3973/14725 [2:03:30<5:19:19,  1.78s/it]
 27%|██▋       | 3974/14725 [2:03:31<5:19:51,  1.79s/it]
 27%|██▋       | 3975/14725 [2:03:33<5:21:32,  1.79s/it]
 27%|██▋       | 3976/14725 [2:03:35<5:19:33,  1.78s/it]
 27%|██▋       | 3977/14725 [2:03:37<5:18:54,  1.78s/it]
 27%|██▋       | 3978/14725 [2:03:38<5:18:08,  1.78s/it]
 27%|██▋       | 3979/14725 [2:03:40<5:23:41,  1.81s/it]
 27%|██▋       | 3980/14725 [2:03:42<5:23:40,  1.81s/it]
                                                        

 27%|██▋       | 3980/14725 [2:03:42<5:23:40,  1.81s/it]
 27%|██▋       | 3981/14725 [2:03:44<5:23:09,  1.80s/it]
 27%|██▋       | 3982/14725 [2:03:46<5:21:50,  1.80s/it]
 27%|██▋       | 3983/14725 [2:03:48<5:19:35,  1.79s/it]
 27%|��█▋       | 3984/14725 [2:03:49<5:20:45,  1.79s/it]
 27%|██▋       | 3985/14725 [2:03:51<5:20:55,  1.79s/it]
 27%|██▋       | 3986/14725 [2:03:53<5:18:52,  1.78s/it]
 27%|██▋       | 3987/14725 [2:03:55<5
+0: {'loss': 0.8454, 'grad_norm': 0.6928466935975555, 'learning_rate': 2e-05, 'epoch': 0.27}
+0: {'loss': 0.8374, 'grad_norm': 0.7067910784463819, 'learning_rate': 2e-05, 'epoch': 0.27}
+0: :33:45,  1.86s/it]
 27%|██▋       | 3988/14725 [2:03:57<5:32:51,  1.86s/it]
 27%|██▋       | 3989/14725 [2:03:59<5:29:03,  1.84s/it]
 27%|██▋       | 3990/14725 [2:04:00<5:24:11,  1.81s/it]
                                                        

 27%|██▋       | 3990/14725 [2:04:00<5:24:11,  1.81s/it]
 27%|██▋       | 3991/14725 [2:04:02<5:21:16,  1.80s/it]
 27%|██▋       | 3992/14725 [2:04:04<5:20:50,  1.79s/it]
 27%|██▋       | 3993/14725 [2:04:06<5:26:36,  1.83s/it]
 27%|██▋       | 3994/14725 [2:04:08<5:25:41,  1.82s/it]
 27%|██▋       | 3995/14725 [2:04:09<5:26:08,  1.82s/it]
 27%|██▋       | 3996/14725 [2:04:11<5:24:33,  1.81s/it]
 27%|██▋       | 3997/14725 [2:04:13<5:21:56,  1.80s/it]
 27%|██▋       | 3998/14725 [2:04:15<5:24:02,  1.81s/it]
 27%|██▋       | 3999/14725 [2:04:17<5:21:58,  1.80s/it]
 27%|██▋       | 4000/14725 [2:04:18<5:19:56,  1.79s/it]
                                                        

 27%|�
+0: {'loss': 0.8643, 'grad_norm': 0.7593462348272969, 'learning_rate': 2e-05, 'epoch': 0.27}
+0: �█▋       | 4000/14725 [2:04:18<5:19:56,  1.79s/it]
 27%|██▋       | 4001/14725 [2:04:20<5:19:36,  1.79s/it]
 27%|██▋       | 4002/14725 [2:04:22<5:20:42,  1.79s/it]
 27%|██▋       | 4003/14725 [2:04:24<5:22:03,  1.80s/it]
 27%|██▋       | 4004/14725 [2:04:26<5:22:41,  1.81s/it]
 27%|██▋       | 4005/14725 [2:04:27<5:21:37,  1.80s/it]
 27%|██▋       | 4006/14725 [2:04:29<5:23:04,  1.81s/it]
 27%|██▋       | 4007/14725 [2:04:31<5:24:42,  1.82s/it]
 27%|██▋       | 4008/14725 [2:04:33<5:27:16,  1.83s/it]
 27%|██▋       | 4009/14725 [2:04:35<5:26:48,  1.83s/it]
 27%|██▋       | 4010/14725 [2:04:37<5:24:41,  1.82s/it]
                                                        

 27%|██▋       | 4010/14725 [2:04:37<5:24:41,  1.82s/it]
 27%|██▋       | 4011/14725 [2:04:38<5:26:03,  1.83s/it]
 27%|██▋       | 4012/14725 [2:04:40<5:25:04,  1.82s/it]
 27%|██▋       | 4013/14725 [2:04:42<5:24:05,  1.82s/it]
 27%|██▋       | 4014/
+0: {'loss': 0.844, 'grad_norm': 0.777045457940314, 'learning_rate': 2e-05, 'epoch': 0.27}
+0: 14725 [2:04:44<5:25:30,  1.82s/it]
 27%|██▋       | 4015/14725 [2:04:46<5:29:26,  1.85s/it]
 27%|██▋       | 4016/14725 [2:04:48<5:27:02,  1.83s/it]
 27%|██▋       | 4017/14725 [2:04:49<5:28:02,  1.84s/it]
 27%|██▋       | 4018/14725 [2:04:51<5:26:33,  1.83s/it]
 27%|██▋       | 4019/14725 [2:04:53<5:25:07,  1.82s/it]
 27%|██▋       | 4020/14725 [2:04:55<5:30:35,  1.85s/it]
                                                        

 27%|██▋       | 4020/14725 [2:04:55<5:30:35,  1.85s/it]
 27%|██▋       | 4021/14725 [2:04:57<5:32:00,  1.86s/it]
 27%|██▋       | 4022/14725 [2:04:59<5:29:32,  1.85s/it]
 27%|██▋       | 4023/14725 [2:05:00<5:28:26,  1.84s/it]
 27%|██▋       | 4024/14725 [2:05:02<5:26:15,  1.83s/it]
 27%|██▋       | 4025/14725 [2:05:04<5:24:58,  1.82s/it]
 27%|██▋       | 4026/14725 [2:05:06<5:21:42,  1.80s/it]
 27%|██▋       | 4027/14725 [2:05:08<5:27:15,  1.84s/it]
 27%|██▋       | 4028/14725 [2:05:10<5:27:2
+0: {'loss': 0.8679, 'grad_norm': 0.8178341695248973, 'learning_rate': 2e-05, 'epoch': 0.27}
+0: {'loss': 0.88, 'grad_norm': 0.7932703673136704, 'learning_rate': 2e-05, 'epoch': 0.27}
+0: 2,  1.84s/it]
 27%|██▋       | 4029/14725 [2:05:11<5:24:11,  1.82s/it]
 27%|██▋       | 4030/14725 [2:05:13<5:22:06,  1.81s/it]
                                                        

 27%|██▋       | 4030/14725 [2:05:13<5:22:06,  1.81s/it]
 27%|██▋       | 4031/14725 [2:05:15<5:20:28,  1.80s/it]
 27%|██▋       | 4032/14725 [2:05:17<5:22:14,  1.81s/it]
 27%|██▋       | 4033/14725 [2:05:18<5:19:14,  1.79s/it]
 27%|██▋       | 4034/14725 [2:05:20<5:17:27,  1.78s/it]
 27%|██▋       | 4035/14725 [2:05:22<5:16:30,  1.78s/it]
 27%|██▋       | 4036/14725 [2:05:24<5:21:34,  1.81s/it]
 27%|██▋       | 4037/14725 [2:05:26<5:22:53,  1.81s/it]
 27%|██▋       | 4038/14725 [2:05:28<5:23:59,  1.82s/it]
 27%|██▋       | 4039/14725 [2:05:29<5:24:33,  1.82s/it]
 27%|██▋       | 4040/14725 [2:05:31<5:25:20,  1.83s/it]
                                                        

 27%|██▋       | 4040/14725 [2:05:31<5:25:20,  1.83s/it]
 27%|██�
+0: {'loss': 0.8571, 'grad_norm': 0.7760996074571961, 'learning_rate': 2e-05, 'epoch': 0.28}
+0: ��       | 4041/14725 [2:05:33<5:29:36,  1.85s/it]
 27%|██▋       | 4042/14725 [2:05:35<5:26:48,  1.84s/it]
 27%|██���       | 4043/14725 [2:05:37<5:22:36,  1.81s/it]
 27%|██▋       | 4044/14725 [2:05:38<5:22:01,  1.81s/it]
 27%|██▋       | 4045/14725 [2:05:40<5:21:46,  1.81s/it]
 27%|██▋       | 4046/14725 [2:05:42<5:19:36,  1.80s/it]
 27%|██▋       | 4047/14725 [2:05:44<5:19:43,  1.80s/it]
 27%|██▋       | 4048/14725 [2:05:46<5:23:05,  1.82s/it]
 27%|██▋       | 4049/14725 [2:05:48<5:25:04,  1.83s/it]
 28%|██▊       | 4050/14725 [2:05:49<5:26:36,  1.84s/it]
                                                        

 28%|██▊       | 4050/14725 [2:05:49<5:26:36,  1.84s/it]
 28%|██▊       | 4051/14725 [2:05:51<5:25:32,  1.83s/it]
 28%|██▊       | 4052/14725 [2:05:53<5:24:13,  1.82s/it]
 28%|██▊       | 4053/14725 [2:05:55<5:24:34,  1.82s/it]
 28%|██▊       | 4054/14725 [2:05:57<5:22:15,  1.81s/it]
 28%|██▊       | 4055/14725
+0: {'loss': 0.8567, 'grad_norm': 0.8291454634869293, 'learning_rate': 2e-05, 'epoch': 0.28}
+0:  [2:05:58<5:21:09,  1.81s/it]
 28%|██▊       | 4056/14725 [2:06:00<5:21:55,  1.81s/it]
 28%|██▊       | 4057/14725 [2:06:02<5:19:49,  1.80s/it]
 28%|██▊       | 4058/14725 [2:06:04<5:21:37,  1.81s/it]
 28%|██▊       | 4059/14725 [2:06:06<5:23:09,  1.82s/it]
 28%|██▊       | 4060/14725 [2:06:07<5:22:38,  1.82s/it]
                                                        

 28%|██▊       | 4060/14725 [2:06:07<5:22:38,  1.82s/it]
 28%|██▊       | 4061/14725 [2:06:09<5:24:48,  1.83s/it]
 28%|██▊       | 4062/14725 [2:06:11<5:22:12,  1.81s/it]
 28%|██▊       | 4063/14725 [2:06:13<5:22:57,  1.82s/it]
 28%|██▊       | 4064/14725 [2:06:15<5:23:52,  1.82s/it]
 28%|██▊       | 4065/14725 [2:06:17<5:25:40,  1.83s/it]
 28%|██▊       | 4066/14725 [2:06:18<5:22:40,  1.82s/it]
 28%|██▊       | 4067/14725 [2:06:20<5:21:59,  1.81s/it]
 28%|██▊       | 4068/14725 [2:06:22<5:19:50,  1.80s/it]
 28%|██▊       | 4069/14725 [2:06:24<5:16:51,  1
+0: {'loss': 0.8561, 'grad_norm': 0.7784873082333769, 'learning_rate': 2e-05, 'epoch': 0.28}
+0: {'loss': 0.8526, 'grad_norm': 0.7775661371672333, 'learning_rate': 2e-05, 'epoch': 0.28}
+0: .78s/it]
 28%|██▊       | 4070/14725 [2:06:26<5:20:12,  1.80s/it]
                                                        

 28%|██▊       | 4070/14725 [2:06:26<5:20:12,  1.80s/it]
 28%|██▊       | 4071/14725 [2:06:27<5:22:54,  1.82s/it]
 28%|██▊       | 4072/14725 [2:06:29<5:23:49,  1.82s/it]
 28%|██▊       | 4073/14725 [2:06:31<5:27:28,  1.84s/it]
 28%|██▊       | 4074/14725 [2:06:33<5:23:55,  1.82s/it]
 28%|██▊       | 4075/14725 [2:06:35<5:21:38,  1.81s/it]
 28%|██▊       | 4076/14725 [2:06:37<5:20:32,  1.81s/it]
 28%|██▊       | 4077/14725 [2:06:38<5:16:56,  1.79s/it]
 28%|██▊       | 4078/14725 [2:06:40<5:15:59,  1.78s/it]
 28%|██▊       | 4079/14725 [2:06:42<5:18:18,  1.79s/it]
 28%|██▊       | 4080/14725 [2:06:44<5:20:58,  1.81s/it]
                                                        

 28%|██▊       | 4080/14725 [2:06:44<5:20:58,  1.81s/it]
 28%|██▊       | 4081/14725 [2:06:46<5:34:20,  1.88s/it]
 28%|██▊   
+0: {'loss': 0.8538, 'grad_norm': 0.7632268151152438, 'learning_rate': 2e-05, 'epoch': 0.28}
+0:     | 4082/14725 [2:06:48<5:33:42,  1.88s/it]
 28%|██▊       | 4083/14725 [2:06:49<5:31:05,  1.87s/it]
 28%|██▊       | 4084/14725 [2:06:51<5:29:36,  1.86s/it]
 28%|██▊       | 4085/14725 [2:06:53<5:34:30,  1.89s/it]
 28%|██▊       | 4086/14725 [2:06:55<5:33:04,  1.88s/it]
 28%|██▊       | 4087/14725 [2:06:57<5:26:20,  1.84s/it]
 28%|██▊       | 4088/14725 [2:06:59<5:26:23,  1.84s/it]
 28%|██▊       | 4089/14725 [2:07:01<5:23:33,  1.83s/it]
 28%|██▊       | 4090/14725 [2:07:02<5:20:34,  1.81s/it]
                                                        

 28%|██▊       | 4090/14725 [2:07:02<5:20:34,  1.81s/it]
 28%|██▊       | 4091/14725 [2:07:04<5:18:38,  1.80s/it]
 28%|██▊       | 4092/14725 [2:07:06<5:18:25,  1.80s/it]
 28%|██▊       | 4093/14725 [2:07:08<5:19:12,  1.80s/it]
 28%|██▊       | 4094/14725 [2:07:09<5:21:07,  1.81s/it]
 28%|██▊       | 4095/14725 [2:07:11<5:19:22,  1.80s/it]
 28%|██▊       | 4096/14725 [2:0
+0: {'loss': 0.8308, 'grad_norm': 0.7474339678934013, 'learning_rate': 2e-05, 'epoch': 0.28}
+0: 7:13<5:21:17,  1.81s/it]
 28%|██▊       | 4097/14725 [2:07:15<5:21:35,  1.82s/it]
 28%|██▊       | 4098/14725 [2:07:17<5:18:52,  1.80s/it]
 28%|██▊       | 4099/14725 [2:07:19<5:19:19,  1.80s/it]
 28%|██▊       | 4100/14725 [2:07:20<5:19:39,  1.81s/it]
                                                        

 28%|██▊       | 4100/14725 [2:07:20<5:19:39,  1.81s/it]
 28%|██▊       | 4101/14725 [2:07:22<5:23:07,  1.82s/it]
 28%|██▊       | 4102/14725 [2:07:24<5:28:16,  1.85s/it]
 28%|██▊       | 4103/14725 [2:07:26<5:24:57,  1.84s/it]
 28%|██▊       | 4104/14725 [2:07:28<5:20:56,  1.81s/it]
 28%|██▊       | 4105/14725 [2:07:29<5:20:06,  1.81s/it]
 28%|██▊       | 4106/14725 [2:07:31<5:19:25,  1.80s/it]
 28%|██▊       | 4107/14725 [2:07:33<5:18:48,  1.80s/it]
 28%|██▊       | 4108/14725 [2:07:35<5:17:38,  1.80s/it]
 28%|██▊       | 4109/14725 [2:07:37<5:17:23,  1.79s/it]
 28%|██▊       | 4110/14725 [2:07:38<5:19:14,  1.80s/
+0: {'loss': 0.8404, 'grad_norm': 0.7797300306178774, 'learning_rate': 2e-05, 'epoch': 0.28}
+0: {'loss': 0.8471, 'grad_norm': 0.8213813653372116, 'learning_rate': 2e-05, 'epoch': 0.28}
+0: it]
                                                        

 28%|██▊       | 4110/14725 [2:07:38<5:19:14,  1.80s/it]
 28%|██▊       | 4111/14725 [2:07:40<5:19:05,  1.80s/it]
 28%|██▊       | 4112/14725 [2:07:42<5:20:20,  1.81s/it]
 28%|██▊       | 4113/14725 [2:07:44<5:17:52,  1.80s/it]
 28%|██▊       | 4114/14725 [2:07:46<5:18:48,  1.80s/it]
 28%|██▊       | 4115/14725 [2:07:47<5:19:13,  1.81s/it]
 28%|██▊       | 4116/14725 [2:07:49<5:16:41,  1.79s/it]
 28%|██▊       | 4117/14725 [2:07:51<5:14:59,  1.78s/it]
 28%|██▊       | 4118/14725 [2:07:53<5:14:39,  1.78s/it]
 28%|██▊       | 4119/14725 [2:07:55<5:17:24,  1.80s/it]
 28%|██▊       | 4120/14725 [2:07:56<5:20:50,  1.82s/it]
                                                        

 28%|██▊       | 4120/14725 [2:07:56<5:20:50,  1.82s/it]
 28%|██▊       | 4121/14725 [2:07:58<5:21:51,  1.82s/it]
 28%|██▊       | 4122/14725 [2:08:00<5:19:29,  1.81s/it]
 28%|██▊       |
+0: {'loss': 0.8721, 'grad_norm': 0.7843559576189751, 'learning_rate': 2e-05, 'epoch': 0.28}
+0:  4123/14725 [2:08:02<5:20:44,  1.82s/it]
 28%|██▊       | 4124/14725 [2:08:04<5:24:55,  1.84s/it]
 28%|██▊       | 4125/14725 [2:08:06<5:21:18,  1.82s/it]
 28%|██▊       | 4126/14725 [2:08:07<5:20:16,  1.81s/it]
 28%|██▊       | 4127/14725 [2:08:09<5:17:36,  1.80s/it]
 28%|██▊       | 4128/14725 [2:08:11<5:19:37,  1.81s/it]
 28%|██▊       | 4129/14725 [2:08:13<5:18:59,  1.81s/it]
 28%|██▊       | 4130/14725 [2:08:15<5:16:02,  1.79s/it]
                                                        

 28%|██▊       | 4130/14725 [2:08:15<5:16:02,  1.79s/it]
 28%|██▊       | 4131/14725 [2:08:16<5:17:18,  1.80s/it]
 28%|██▊       | 4132/14725 [2:08:18<5:20:25,  1.81s/it]
 28%|██▊       | 4133/14725 [2:08:20<5:21:49,  1.82s/it]
 28%|██▊       | 4134/14725 [2:08:22<5:21:14,  1.82s/it]
 28%|██▊       | 4135/14725 [2:08:24<5:18:47,  1.81s/it]
 28%|██▊       | 4136/14725 [2:08:25<5:18:41,  1.81s/it]
 28%|██▊       | 4137/14725 [2:08:27<
+0: {'loss': 0.8553, 'grad_norm': 0.8182952953278645, 'learning_rate': 2e-05, 'epoch': 0.28}
+0: {'loss': 0.8416, 'grad_norm': 0.7393349432734593, 'learning_rate': 2e-05, 'epoch': 0.28}
+0: 5:20:48,  1.82s/it]
 28%|██▊       | 4138/14725 [2:08:29<5:26:02,  1.85s/it]
 28%|██▊       | 4139/14725 [2:08:31<5:23:56,  1.84s/it]
 28%|██▊       | 4140/14725 [2:08:33<5:28:26,  1.86s/it]
                                                        

 28%|██▊       | 4140/14725 [2:08:33<5:28:26,  1.86s/it]
 28%|██▊       | 4141/14725 [2:08:35<5:25:38,  1.85s/it]
 28%|██▊       | 4142/14725 [2:08:36<5:20:16,  1.82s/it]
 28%|██▊       | 4143/14725 [2:08:38<5:16:23,  1.79s/it]
 28%|██▊       | 4144/14725 [2:08:40<5:21:00,  1.82s/it]
 28%|██▊       | 4145/14725 [2:08:42<5:20:20,  1.82s/it]
 28%|██▊       | 4146/14725 [2:08:44<5:19:01,  1.81s/it]
 28%|██▊       | 4147/14725 [2:08:45<5:18:20,  1.81s/it]
 28%|██▊       | 4148/14725 [2:08:47<5:15:47,  1.79s/it]
 28%|██▊       | 4149/14725 [2:08:49<5:18:28,  1.81s/it]
 28%|██▊       | 4150/14725 [2:08:51<5:17:49,  1.80s/it]
                                                        

 28%|�
+0: {'loss': 0.8429, 'grad_norm': 0.7622387957035973, 'learning_rate': 2e-05, 'epoch': 0.28}
+0: ��█▊       | 4150/14725 [2:08:51<5:17:49,  1.80s/it]
 28%|██▊       | 4151/14725 [2:08:53<5:16:31,  1.80s/it]
 28%|██▊       | 4152/14725 [2:08:55<5:19:44,  1.81s/it]
 28%|██▊       | 4153/14725 [2:08:56<5:19:27,  1.81s/it]
 28%|██▊       | 4154/14725 [2:08:58<5:22:37,  1.83s/it]
 28%|██▊       | 4155/14725 [2:09:00<5:24:02,  1.84s/it]
 28%|██▊       | 4156/14725 [2:09:02<5:21:53,  1.83s/it]
 28%|██▊       | 4157/14725 [2:09:04<5:21:38,  1.83s/it]
 28%|██▊       | 4158/14725 [2:09:06<5:27:46,  1.86s/it]
 28%|██▊       | 4159/14725 [2:09:08<5:28:25,  1.86s/it]
 28%|██▊       | 4160/14725 [2:09:09<5:24:13,  1.84s/it]
                                                        

 28%|██▊       | 4160/14725 [2:09:09<5:24:13,  1.84s/it]
 28%|██▊       | 4161/14725 [2:09:11<5:23:43,  1.84s/it]
 28%|██▊       | 4162/14725 [2:09:13<5:22:01,  1.83s/it]
 28%|██▊       | 4163/14725 [2:09:15<5:23:24,  1.84s/it]
 28%|██▊       | 4164
+0: {'loss': 0.8572, 'grad_norm': 0.7738004475012711, 'learning_rate': 2e-05, 'epoch': 0.28}
+0: /14725 [2:09:17<5:23:33,  1.84s/it]
 28%|██▊       | 4165/14725 [2:09:18<5:23:17,  1.84s/it]
 28%|██▊       | 4166/14725 [2:09:20<5:19:16,  1.81s/it]
 28%|██▊       | 4167/14725 [2:09:22<5:24:36,  1.84s/it]
 28%|██▊       | 4168/14725 [2:09:24<5:27:45,  1.86s/it]
 28%|██▊       | 4169/14725 [2:09:26<5:27:05,  1.86s/it]
 28%|██▊       | 4170/14725 [2:09:28<5:21:36,  1.83s/it]
                                                        

 28%|██▊       | 4170/14725 [2:09:28<5:21:36,  1.83s/it]
 28%|██▊       | 4171/14725 [2:09:29<5:19:19,  1.82s/it]
 28%|██▊       | 4172/14725 [2:09:31<5:15:45,  1.80s/it]
 28%|██▊       | 4173/14725 [2:09:33<5:14:28,  1.79s/it]
 28%|██▊       | 4174/14725 [2:09:35<5:18:22,  1.81s/it]
 28%|██▊       | 4175/14725 [2:09:37<5:21:21,  1.83s/it]
 28%|██▊       | 4176/14725 [2:09:38<5:19:34,  1.82s/it]
 28%|██▊       | 4177/14725 [2:09:40<5:19:58,  1.82s/it]
 28%|██▊       | 4178/14725 [2:09:42<5:19:
+0: {'loss': 0.8626, 'grad_norm': 0.878114870797975, 'learning_rate': 2e-05, 'epoch': 0.28}
+0: {'loss': 0.8635, 'grad_norm': 0.8191143362898637, 'learning_rate': 2e-05, 'epoch': 0.28}
+0: 02,  1.81s/it]
 28%|██▊       | 4179/14725 [2:09:44<5:18:58,  1.81s/it]
 28%|██▊       | 4180/14725 [2:09:46<5:27:04,  1.86s/it]
                                                        

 28%|██▊       | 4180/14725 [2:09:46<5:27:04,  1.86s/it]
 28%|██▊       | 4181/14725 [2:09:48<5:24:16,  1.85s/it]
 28%|██▊       | 4182/14725 [2:09:50<5:23:05,  1.84s/it]
 28%|██▊       | 4183/14725 [2:09:51<5:22:12,  1.83s/it]
 28%|██▊       | 4184/14725 [2:09:53<5:18:34,  1.81s/it]
 28%|██▊       | 4185/14725 [2:09:55<5:17:30,  1.81s/it]
 28%|██▊       | 4186/14725 [2:09:57<5:15:41,  1.80s/it]
 28%|██▊       | 4187/14725 [2:09:58<5:14:31,  1.79s/it]
 28%|██▊       | 4188/14725 [2:10:00<5:13:17,  1.78s/it]
 28%|██▊       | 4189/14725 [2:10:02<5:16:22,  1.80s/it]
 28%|██▊       | 4190/14725 [2:10:04<5:15:47,  1.80s/it]
                                                        

 28%|██▊       | 4190/14725 [2:10:04<5:15:47,  1.80s/it]
 28%|██
+0: {'loss': 0.8475, 'grad_norm': 0.8289587426181059, 'learning_rate': 2e-05, 'epoch': 0.29}
+0: ▊       | 4191/14725 [2:10:06<5:16:43,  1.80s/it]
 28%|██▊       | 4192/14725 [2:10:08<5:19:13,  1.82s/it]
 28%|██▊       | 4193/14725 [2:10:09<5:15:53,  1.80s/it]
 28%|██▊       | 4194/14725 [2:10:11<5:14:54,  1.79s/it]
 28%|██▊       | 4195/14725 [2:10:13<5:19:49,  1.82s/it]
 28%|██▊       | 4196/14725 [2:10:15<5:17:45,  1.81s/it]
 29%|██▊       | 4197/14725 [2:10:17<5:18:15,  1.81s/it]
 29%|██▊       | 4198/14725 [2:10:18<5:16:38,  1.80s/it]
 29%|██▊       | 4199/14725 [2:10:20<5:17:50,  1.81s/it]
 29%|██▊       | 4200/14725 [2:10:22<5:15:24,  1.80s/it]
                                                        

 29%|██▊       | 4200/14725 [2:10:22<5:15:24,  1.80s/it]
 29%|██▊       | 4201/14725 [2:10:24<5:13:36,  1.79s/it]
 29%|██▊       | 4202/14725 [2:10:26<5:15:43,  1.80s/it]
 29%|██▊       | 4203/14725 [2:10:27<5:15:04,  1.80s/it]
 29%|██▊       | 4204/14725 [2:10:29<5:16:18,  1.80s/it]
 29%|██▊       | 4205/1472
+0: {'loss': 0.8539, 'grad_norm': 0.788102871707846, 'learning_rate': 2e-05, 'epoch': 0.29}
+0: 5 [2:10:31<5:20:10,  1.83s/it]
 29%|██▊       | 4206/14725 [2:10:33<5:15:38,  1.80s/it]
 29%|██▊       | 4207/14725 [2:10:35<5:14:20,  1.79s/it]
 29%|██▊       | 4208/14725 [2:10:36<5:12:53,  1.79s/it]
 29%|██▊       | 4209/14725 [2:10:38<5:13:19,  1.79s/it]
 29%|██▊       | 4210/14725 [2:10:40<5:15:24,  1.80s/it]
                                                        

 29%|██▊       | 4210/14725 [2:10:40<5:15:24,  1.80s/it]
 29%|██▊       | 4211/14725 [2:10:42<5:17:24,  1.81s/it]
 29%|██▊       | 4212/14725 [2:10:44<5:26:29,  1.86s/it]
 29%|██▊       | 4213/14725 [2:10:46<5:27:12,  1.87s/it]
 29%|██▊       | 4214/14725 [2:10:47<5:24:33,  1.85s/it]
 29%|██▊       | 4215/14725 [2:10:49<5:22:07,  1.84s/it]
 29%|██▊       | 4216/14725 [2:10:51<5:26:50,  1.87s/it]
 29%|██▊       | 4217/14725 [2:10:53<5:25:57,  1.86s/it]
 29%|██▊       | 4218/14725 [2:10:55<5:22:12,  1.84s/it]
 29%|██▊       | 4219/14725 [2:10:57<5:21:05,  
+0: {'loss': 0.8532, 'grad_norm': 0.7516077343096843, 'learning_rate': 2e-05, 'epoch': 0.29}
+0: {'loss': 0.8521, 'grad_norm': 0.8119174557451803, 'learning_rate': 2e-05, 'epoch': 0.29}
+0: 1.83s/it]
 29%|██▊       | 4220/14725 [2:10:58<5:22:56,  1.84s/it]
                                                        

 29%|██▊       | 4220/14725 [2:10:59<5:22:56,  1.84s/it]
 29%|██▊       | 4221/14725 [2:11:00<5:22:17,  1.84s/it]
 29%|██▊       | 4222/14725 [2:11:02<5:20:40,  1.83s/it]
 29%|██▊       | 4223/14725 [2:11:04<5:19:50,  1.83s/it]
 29%|██▊       | 4224/14725 [2:11:06<5:16:04,  1.81s/it]
 29%|██▊       | 4225/14725 [2:11:08<5:16:08,  1.81s/it]
 29%|██▊       | 4226/14725 [2:11:10<5:35:08,  1.92s/it]
 29%|██▊       | 4227/14725 [2:11:11<5:28:34,  1.88s/it]
 29%|██▊       | 4228/14725 [2:11:13<5:28:48,  1.88s/it]
 29%|██▊       | 4229/14725 [2:11:15<5:26:23,  1.87s/it]
 29%|██▊       | 4230/14725 [2:11:17<5:25:55,  1.86s/it]
                                                        

 29%|██▊       | 4230/14725 [2:11:17<5:25:55,  1.86s/it]
 29%|██▊       | 4231/14725 [2:11:19<5:20:47,  1.83s/it]
 29%|██▊  
+0: {'loss': 0.8586, 'grad_norm': 0.8085355849567005, 'learning_rate': 2e-05, 'epoch': 0.29}
+0:      | 4232/14725 [2:11:21<5:17:21,  1.81s/it]
 29%|██▊       | 4233/14725 [2:11:22<5:20:30,  1.83s/it]
 29%|██▉       | 4234/14725 [2:11:24<5:20:37,  1.83s/it]
 29%|██▉       | 4235/14725 [2:11:26<5:17:47,  1.82s/it]
 29%|██▉       | 4236/14725 [2:11:28<5:15:18,  1.80s/it]
 29%|██▉       | 4237/14725 [2:11:30<5:47:46,  1.99s/it]
 29%|██▉       | 4238/14725 [2:11:32<5:36:31,  1.93s/it]
 29%|██▉       | 4239/14725 [2:11:34<5:29:14,  1.88s/it]
 29%|██▉       | 4240/14725 [2:11:36<5:27:46,  1.88s/it]
                                                        

 29%|██▉       | 4240/14725 [2:11:36<5:27:46,  1.88s/it]
 29%|██▉       | 4241/14725 [2:11:38<5:26:14,  1.87s/it]
 29%|██▉       | 4242/14725 [2:11:39<5:21:13,  1.84s/it]
 29%|██▉       | 4243/14725 [2:11:41<5:16:42,  1.81s/it]
 29%|██▉       | 4244/14725 [2:11:43<5:16:50,  1.81s/it]
 29%|██▉       | 4245/14725 [2:11:45<5:16:06,  1.81s/it]
 29%|██▉       | 4246/14725 [2:
+0: {'loss': 0.8325, 'grad_norm': 0.8438612104756338, 'learning_rate': 2e-05, 'epoch': 0.29}
+0: 11:46<5:12:57,  1.79s/it]
 29%|██▉       | 4247/14725 [2:11:48<5:16:13,  1.81s/it]
 29%|██▉       | 4248/14725 [2:11:50<5:17:33,  1.82s/it]
 29%|██▉       | 4249/14725 [2:11:52<5:16:01,  1.81s/it]
 29%|██▉       | 4250/14725 [2:11:54<5:15:24,  1.81s/it]
                                                        

 29%|██▉       | 4250/14725 [2:11:54<5:15:24,  1.81s/it]
 29%|██▉       | 4251/14725 [2:11:56<5:14:21,  1.80s/it]
 29%|██▉       | 4252/14725 [2:11:57<5:18:15,  1.82s/it]
 29%|██▉       | 4253/14725 [2:11:59<5:15:36,  1.81s/it]
 29%|██▉       | 4254/14725 [2:12:01<5:16:17,  1.81s/it]
 29%|██▉       | 4255/14725 [2:12:03<5:12:54,  1.79s/it]
 29%|██▉       | 4256/14725 [2:12:04<5:11:07,  1.78s/it]
 29%|██▉       | 4257/14725 [2:12:06<5:15:21,  1.81s/it]
 29%|██▉       | 4258/14725 [2:12:08<5:13:07,  1.79s/it]
 29%|██▉       | 4259/14725 [2:12:10<5:14:21,  1.80s/it]
 29%|██▉       | 4260/14725 [2:12:12<5:15:31,  1.81s
+0: {'loss': 0.8449, 'grad_norm': 0.7586653783272053, 'learning_rate': 2e-05, 'epoch': 0.29}
+0: {'loss': 0.8451, 'grad_norm': 0.8112043683633804, 'learning_rate': 2e-05, 'epoch': 0.29}
+0: /it]
                                                        

 29%|██▉       | 4260/14725 [2:12:12<5:15:31,  1.81s/it]
 29%|██▉       | 4261/14725 [2:12:14<5:21:04,  1.84s/it]
 29%|██▉       | 4262/14725 [2:12:16<5:22:53,  1.85s/it]
 29%|██▉       | 4263/14725 [2:12:17<5:24:17,  1.86s/it]
 29%|██▉       | 4264/14725 [2:12:19<5:21:46,  1.85s/it]
 29%|██▉       | 4265/14725 [2:12:21<5:22:20,  1.85s/it]
 29%|██▉       | 4266/14725 [2:12:23<5:23:20,  1.85s/it]
 29%|██▉       | 4267/14725 [2:12:25<5:18:11,  1.83s/it]
 29%|██▉       | 4268/14725 [2:12:27<5:21:12,  1.84s/it]
 29%|██▉       | 4269/14725 [2:12:28<5:16:47,  1.82s/it]
 29%|██▉       | 4270/14725 [2:12:30<5:12:50,  1.80s/it]
                                                        

 29%|██▉       | 4270/14725 [2:12:30<5:12:50,  1.80s/it]
 29%|██▉       | 4271/14725 [2:12:32<5:14:48,  1.81s/it]
 29%|██▉       | 4272/14725 [2:12:34<5:18:06,  1.83s/it]
 29%|██▉       
+0: {'loss': 0.8514, 'grad_norm': 0.7712588664243292, 'learning_rate': 2e-05, 'epoch': 0.29}
+0: | 4273/14725 [2:12:36<5:17:21,  1.82s/it]
 29%|██▉       | 4274/14725 [2:12:37<5:15:28,  1.81s/it]
 29%|██▉       | 4275/14725 [2:12:39<5:15:07,  1.81s/it]
 29%|██▉       | 4276/14725 [2:12:41<5:12:11,  1.79s/it]
 29%|██▉       | 4277/14725 [2:12:43<5:12:05,  1.79s/it]
 29%|██▉       | 4278/14725 [2:12:45<5:11:48,  1.79s/it]
 29%|██▉       | 4279/14725 [2:12:46<5:13:45,  1.80s/it]
 29%|██▉       | 4280/14725 [2:12:48<5:13:47,  1.80s/it]
                                                        

 29%|██▉       | 4280/14725 [2:12:48<5:13:47,  1.80s/it]
 29%|██▉       | 4281/14725 [2:12:50<5:11:45,  1.79s/it]
 29%|██▉       | 4282/14725 [2:12:52<5:12:04,  1.79s/it]
 29%|██▉       | 4283/14725 [2:12:54<5:13:43,  1.80s/it]
 29%|██▉       | 4284/14725 [2:12:55<5:11:31,  1.79s/it]
 29%|██▉       | 4285/14725 [2:12:57<5:12:12,  1.79s/it]
 29%|██▉       | 4286/14725 [2:12:59<5:10:45,  1.79s/it]
 29%|██▉       | 4287/14725 [2:13:01
+0: {'loss': 0.8472, 'grad_norm': 0.7127348296854689, 'learning_rate': 2e-05, 'epoch': 0.29}
+0: {'loss': 0.8332, 'grad_norm': 0.7941081436472738, 'learning_rate': 2e-05, 'epoch': 0.29}
+0: <5:11:21,  1.79s/it]
 29%|██▉       | 4288/14725 [2:13:02<5:11:21,  1.79s/it]
 29%|██▉       | 4289/14725 [2:13:04<5:11:22,  1.79s/it]
 29%|██▉       | 4290/14725 [2:13:06<5:10:06,  1.78s/it]
                                                        

 29%|██▉       | 4290/14725 [2:13:06<5:10:06,  1.78s/it]
 29%|██▉       | 4291/14725 [2:13:08<5:42:08,  1.97s/it]
 29%|██▉       | 4292/14725 [2:13:10<5:34:10,  1.92s/it]
 29%|██▉       | 4293/14725 [2:13:12<5:27:48,  1.89s/it]
 29%|██▉       | 4294/14725 [2:13:14<5:23:49,  1.86s/it]
 29%|██▉       | 4295/14725 [2:13:16<5:20:09,  1.84s/it]
 29%|██▉       | 4296/14725 [2:13:17<5:16:27,  1.82s/it]
 29%|██▉       | 4297/14725 [2:13:19<5:15:32,  1.82s/it]
 29%|██▉       | 4298/14725 [2:13:21<5:16:27,  1.82s/it]
 29%|██▉       | 4299/14725 [2:13:23<5:16:10,  1.82s/it]
 29%|██▉       | 4300/14725 [2:13:25<5:13:47,  1.81s/it]
                                                        

 29%|
+0: {'loss': 0.8199, 'grad_norm': 0.8193218626222454, 'learning_rate': 2e-05, 'epoch': 0.29}
+0: ██▉       | 4300/14725 [2:13:25<5:13:47,  1.81s/it]
 29%|██▉       | 4301/14725 [2:13:27<5:15:40,  1.82s/it]
 29%|██▉       | 4302/14725 [2:13:29<5:35:34,  1.93s/it]
 29%|██▉       | 4303/14725 [2:13:31<5:45:05,  1.99s/it]
 29%|██▉       | 4304/14725 [2:13:33<5:33:10,  1.92s/it]
 29%|██▉       | 4305/14725 [2:13:34<5:27:35,  1.89s/it]
 29%|██▉       | 4306/14725 [2:13:36<5:19:44,  1.84s/it]
 29%|██▉       | 4307/14725 [2:13:38<5:17:15,  1.83s/it]
 29%|██▉       | 4308/14725 [2:13:40<5:12:44,  1.80s/it]
 29%|██▉       | 4309/14725 [2:13:41<5:11:57,  1.80s/it]
 29%|██▉       | 4310/14725 [2:13:43<5:09:14,  1.78s/it]
                                                        

 29%|██▉       | 4310/14725 [2:13:43<5:09:14,  1.78s/it]
 29%|██▉       | 4311/14725 [2:13:45<5:09:29,  1.78s/it]
 29%|██▉       | 4312/14725 [2:13:47<5:08:16,  1.78s/it]
 29%|██▉       | 4313/14725 [2:13:49<5:22:11,  1.86s/it]
 29%|██▉       | 431
+0: {'loss': 0.8594, 'grad_norm': 0.7368925938614469, 'learning_rate': 2e-05, 'epoch': 0.29}
+0: 4/14725 [2:13:51<5:21:37,  1.85s/it]
 29%|██▉       | 4315/14725 [2:13:52<5:20:21,  1.85s/it]
 29%|██▉       | 4316/14725 [2:13:54<5:16:08,  1.82s/it]
 29%|██▉       | 4317/14725 [2:13:56<5:12:58,  1.80s/it]
 29%|██▉       | 4318/14725 [2:13:58<5:27:22,  1.89s/it]
 29%|██▉       | 4319/14725 [2:14:00<5:24:44,  1.87s/it]
 29%|██▉       | 4320/14725 [2:14:02<5:21:36,  1.85s/it]
                                                        

 29%|██▉       | 4320/14725 [2:14:02<5:21:36,  1.85s/it]
 29%|██▉       | 4321/14725 [2:14:04<5:20:39,  1.85s/it]
 29%|██▉       | 4322/14725 [2:14:05<5:17:56,  1.83s/it]
 29%|██▉       | 4323/14725 [2:14:07<5:18:18,  1.84s/it]
 29%|██▉       | 4324/14725 [2:14:09<5:18:15,  1.84s/it]
 29%|██▉       | 4325/14725 [2:14:11<5:17:46,  1.83s/it]
 29%|██▉       | 4326/14725 [2:14:13<5:23:02,  1.86s/it]
 29%|██▉       | 4327/14725 [2:14:15<5:17:15,  1.83s/it]
 29%|██▉       | 4328/14725 [2:14:16<5:17
+0: {'loss': 0.8419, 'grad_norm': 0.7506298076331483, 'learning_rate': 2e-05, 'epoch': 0.29}
+0: {'loss': 0.8532, 'grad_norm': 0.8155947762557588, 'learning_rate': 2e-05, 'epoch': 0.29}
+0: :03,  1.83s/it]
 29%|██▉       | 4329/14725 [2:14:18<5:18:05,  1.84s/it]
 29%|██▉       | 4330/14725 [2:14:20<5:16:02,  1.82s/it]
                                                        

 29%|██▉       | 4330/14725 [2:14:20<5:16:02,  1.82s/it]
 29%|██▉       | 4331/14725 [2:14:22<5:14:24,  1.81s/it]
 29%|██▉       | 4332/14725 [2:14:24<5:15:56,  1.82s/it]
 29%|██▉       | 4333/14725 [2:14:25<5:14:20,  1.81s/it]
 29%|██▉       | 4334/14725 [2:14:27<5:14:06,  1.81s/it]
 29%|██▉       | 4335/14725 [2:14:29<5:11:42,  1.80s/it]
 29%|██▉       | 4336/14725 [2:14:31<5:11:31,  1.80s/it]
 29%|██▉       | 4337/14725 [2:14:33<5:15:19,  1.82s/it]
 29%|██▉       | 4338/14725 [2:14:34<5:12:30,  1.81s/it]
 29%|██▉       | 4339/14725 [2:14:36<5:13:02,  1.81s/it]
 29%|██▉       | 4340/14725 [2:14:38<5:11:50,  1.80s/it]
                                                        

 29%|██▉       | 4340/14725 [2:14:38<5:11:50,  1.80s/it]
 29%|█�
+0: {'loss': 0.823, 'grad_norm': 0.8389157999191172, 'learning_rate': 2e-05, 'epoch': 0.3}
+0: �▉       | 4341/14725 [2:14:40<5:09:00,  1.79s/it]
 29%|██▉       | 4342/14725 [2:14:42<5:08:52,  1.78s/it]
 29%|██▉       | 4343/14725 [2:14:43<5:14:20,  1.82s/it]
 30%|██▉       | 4344/14725 [2:14:45<5:13:15,  1.81s/it]
 30%|██▉       | 4345/14725 [2:14:47<5:10:24,  1.79s/it]
 30%|██▉       | 4346/14725 [2:14:49<5:10:11,  1.79s/it]
 30%|██▉       | 4347/14725 [2:14:51<5:14:22,  1.82s/it]
 30%|██▉       | 4348/14725 [2:14:53<5:15:21,  1.82s/it]
 30%|██▉       | 4349/14725 [2:14:54<5:18:34,  1.84s/it]
 30%|██▉       | 4350/14725 [2:14:56<5:16:41,  1.83s/it]
                                                        

 30%|██▉       | 4350/14725 [2:14:56<5:16:41,  1.83s/it]
 30%|██▉       | 4351/14725 [2:14:58<5:14:27,  1.82s/it]
 30%|██▉       | 4352/14725 [2:15:00<5:14:29,  1.82s/it]
 30%|██▉       | 4353/14725 [2:15:02<5:13:40,  1.81s/it]
 30%|██▉       | 4354/14725 [2:15:03<5:14:33,  1.82s/it]
 30%|██▉       | 4355/147
+0: {'loss': 0.8554, 'grad_norm': 0.7931121499523536, 'learning_rate': 2e-05, 'epoch': 0.3}
+0: 25 [2:15:05<5:14:34,  1.82s/it]
 30%|██▉       | 4356/14725 [2:15:07<5:15:42,  1.83s/it]
 30%|██▉       | 4357/14725 [2:15:09<5:16:06,  1.83s/it]
 30%|██▉       | 4358/14725 [2:15:11<5:13:51,  1.82s/it]
 30%|██▉       | 4359/14725 [2:15:13<5:15:24,  1.83s/it]
 30%|██▉       | 4360/14725 [2:15:14<5:14:02,  1.82s/it]
                                                        

 30%|██▉       | 4360/14725 [2:15:14<5:14:02,  1.82s/it]
 30%|██▉       | 4361/14725 [2:15:16<5:13:44,  1.82s/it]
 30%|██▉       | 4362/14725 [2:15:18<5:11:54,  1.81s/it]
 30%|██▉       | 4363/14725 [2:15:20<5:12:51,  1.81s/it]
 30%|██▉       | 4364/14725 [2:15:22<5:12:06,  1.81s/it]
 30%|██▉       | 4365/14725 [2:15:23<5:08:56,  1.79s/it]
 30%|██▉       | 4366/14725 [2:15:25<5:07:02,  1.78s/it]
 30%|██▉       | 4367/14725 [2:15:27<5:06:49,  1.78s/it]
 30%|██▉       | 4368/14725 [2:15:29<5:07:03,  1.78s/it]
 30%|██▉       | 4369/14725 [2:15:30<5:05:45, 
+0: {'loss': 0.8545, 'grad_norm': 0.801264608456891, 'learning_rate': 2e-05, 'epoch': 0.3}
+0: {'loss': 0.8475, 'grad_norm': 0.7982781201025753, 'learning_rate': 2e-05, 'epoch': 0.3}
+0:  1.77s/it]
 30%|██▉       | 4370/14725 [2:15:32<5:08:00,  1.78s/it]
                                                        

 30%|██▉       | 4370/14725 [2:15:32<5:08:00,  1.78s/it]
 30%|██▉       | 4371/14725 [2:15:34<5:10:30,  1.80s/it]
 30%|██▉       | 4372/14725 [2:15:36<5:10:15,  1.80s/it]
 30%|██▉       | 4373/14725 [2:15:38<5:09:09,  1.79s/it]
 30%|██▉       | 4374/14725 [2:15:39<5:08:20,  1.79s/it]
 30%|██▉       | 4375/14725 [2:15:41<5:09:58,  1.80s/it]
 30%|██▉       | 4376/14725 [2:15:43<5:13:57,  1.82s/it]
 30%|██▉       | 4377/14725 [2:15:45<5:10:08,  1.80s/it]
 30%|██▉       | 4378/14725 [2:15:47<5:12:12,  1.81s/it]
 30%|██▉       | 4379/14725 [2:15:49<5:11:46,  1.81s/it]
 30%|██▉       | 4380/14725 [2:15:50<5:13:27,  1.82s/it]
                                                        

 30%|██▉       | 4380/14725 [2:15:50<5:13:27,  1.82s/it]
 30%|██▉       | 4381/14725 [2:15:52<5:10:35,  1.80s/it]
 30%|██▉ 
+0: {'loss': 0.8485, 'grad_norm': 0.8149475670452133, 'learning_rate': 2e-05, 'epoch': 0.3}
+0:       | 4382/14725 [2:15:54<5:08:31,  1.79s/it]
 30%|██▉       | 4383/14725 [2:15:56<5:09:54,  1.80s/it]
 30%|██▉       | 4384/14725 [2:15:57<5:08:54,  1.79s/it]
 30%|██▉       | 4385/14725 [2:15:59<5:07:17,  1.78s/it]
 30%|██▉       | 4386/14725 [2:16:01<5:09:29,  1.80s/it]
 30%|██▉       | 4387/14725 [2:16:03<5:11:42,  1.81s/it]
 30%|██▉       | 4388/14725 [2:16:05<5:10:07,  1.80s/it]
 30%|██▉       | 4389/14725 [2:16:06<5:10:02,  1.80s/it]
 30%|██▉       | 4390/14725 [2:16:08<5:07:16,  1.78s/it]
                                                        

 30%|██▉       | 4390/14725 [2:16:08<5:07:16,  1.78s/it]
 30%|██▉       | 4391/14725 [2:16:10<5:05:32,  1.77s/it]
 30%|██▉       | 4392/14725 [2:16:12<5:08:20,  1.79s/it]
 30%|██▉       | 4393/14725 [2:16:14<5:10:09,  1.80s/it]
 30%|██▉       | 4394/14725 [2:16:15<5:11:10,  1.81s/it]
 30%|██▉       | 4395/14725 [2:16:17<5:20:20,  1.86s/it]
 30%|██▉       | 4396/14725 [2
+0: {'loss': 0.8444, 'grad_norm': 0.7575059022609654, 'learning_rate': 2e-05, 'epoch': 0.3}
+0: :16:19<5:17:35,  1.84s/it]
 30%|██▉       | 4397/14725 [2:16:21<5:13:47,  1.82s/it]
 30%|██▉       | 4398/14725 [2:16:23<5:09:28,  1.80s/it]
 30%|██▉       | 4399/14725 [2:16:25<5:10:46,  1.81s/it]
 30%|██▉       | 4400/14725 [2:16:26<5:11:33,  1.81s/it]
                                                        

 30%|██▉       | 4400/14725 [2:16:26<5:11:33,  1.81s/it]
 30%|██▉       | 4401/14725 [2:16:28<5:09:19,  1.80s/it]
 30%|██▉       | 4402/14725 [2:16:30<5:11:18,  1.81s/it]
 30%|██▉       | 4403/14725 [2:16:32<5:10:23,  1.80s/it]
 30%|██▉       | 4404/14725 [2:16:34<5:06:40,  1.78s/it]
 30%|██▉       | 4405/14725 [2:16:35<5:09:47,  1.80s/it]
 30%|██▉       | 4406/14725 [2:16:37<5:09:04,  1.80s/it]
 30%|██▉       | 4407/14725 [2:16:39<5:09:58,  1.80s/it]
 30%|���█▉       | 4408/14725 [2:16:41<5:11:28,  1.81s/it]
 30%|██▉       | 4409/14725 [2:16:43<5:10:14,  1.80s/it]
 30%|██▉       | 4410/14725 [2:16:44<5:11:01,  1.81
+0: {'loss': 0.8577, 'grad_norm': 0.79810878141043, 'learning_rate': 2e-05, 'epoch': 0.3}
+0: {'loss': 0.8507, 'grad_norm': 0.8537997437005391, 'learning_rate': 2e-05, 'epoch': 0.3}
+0: s/it]
                                                        

 30%|██▉       | 4410/14725 [2:16:44<5:11:01,  1.81s/it]
 30%|██▉       | 4411/14725 [2:16:46<5:09:06,  1.80s/it]
 30%|██▉       | 4412/14725 [2:16:48<5:06:06,  1.78s/it]
 30%|██▉       | 4413/14725 [2:16:50<5:04:44,  1.77s/it]
 30%|██▉       | 4414/14725 [2:16:52<5:06:05,  1.78s/it]
 30%|██▉       | 4415/14725 [2:16:53<5:05:36,  1.78s/it]
 30%|██▉       | 4416/14725 [2:16:55<5:08:17,  1.79s/it]
 30%|██▉       | 4417/14725 [2:16:57<5:07:16,  1.79s/it]
 30%|███       | 4418/14725 [2:16:59<5:09:29,  1.80s/it]
 30%|███       | 4419/14725 [2:17:00<5:07:53,  1.79s/it]
 30%|███       | 4420/14725 [2:17:02<5:14:50,  1.83s/it]
                                                        

 30%|███       | 4420/14725 [2:17:02<5:14:50,  1.83s/it]
 30%|███       | 4421/14725 [2:17:04<5:12:52,  1.82s/it]
 30%|███       | 4422/14725 [2:17:06<5:09:22,  1.80s/it]
 30%|███      
+0: {'loss': 0.8364, 'grad_norm': 0.783492083087105, 'learning_rate': 2e-05, 'epoch': 0.3}
+0:  | 4423/14725 [2:17:08<5:10:12,  1.81s/it]
 30%|███       | 4424/14725 [2:17:10<5:09:23,  1.80s/it]
 30%|███       | 4425/14725 [2:17:11<5:09:28,  1.80s/it]
 30%|███       | 4426/14725 [2:17:13<5:09:37,  1.80s/it]
 30%|███       | 4427/14725 [2:17:15<5:07:11,  1.79s/it]
 30%|███       | 4428/14725 [2:17:17<5:05:46,  1.78s/it]
 30%|███       | 4429/14725 [2:17:18<5:04:45,  1.78s/it]
 30%|███       | 4430/14725 [2:17:20<5:04:30,  1.77s/it]
                                                        

 30%|███       | 4430/14725 [2:17:20<5:04:30,  1.77s/it]
 30%|███       | 4431/14725 [2:17:22<5:07:24,  1.79s/it]
 30%|███       | 4432/14725 [2:17:24<5:09:14,  1.80s/it]
 30%|███       | 4433/14725 [2:17:26<5:09:18,  1.80s/it]
 30%|███       | 4434/14725 [2:17:28<5:15:44,  1.84s/it]
 30%|███       | 4435/14725 [2:17:29<5:11:47,  1.82s/it]
 30%|███       | 4436/14725 [2:17:31<5:10:42,  1.81s/it]
 30%|███       | 4437/14725 [2:17:3
+0: {'loss': 0.8475, 'grad_norm': 0.7047520508562654, 'learning_rate': 2e-05, 'epoch': 0.3}
+0: {'loss': 0.8486, 'grad_norm': 0.8051969055023906, 'learning_rate': 2e-05, 'epoch': 0.3}
+0: 3<5:08:07,  1.80s/it]
 30%|███       | 4438/14725 [2:17:35<5:06:18,  1.79s/it]
 30%|███       | 4439/14725 [2:17:37<5:08:25,  1.80s/it]
 30%|███       | 4440/14725 [2:17:38<5:10:02,  1.81s/it]
                                                        

 30%|███       | 4440/14725 [2:17:38<5:10:02,  1.81s/it]
 30%|███       | 4441/14725 [2:17:40<5:07:43,  1.80s/it]
 30%|███       | 4442/14725 [2:17:42<5:12:02,  1.82s/it]
 30%|███       | 4443/14725 [2:17:44<5:10:07,  1.81s/it]
 30%|███       | 4444/14725 [2:17:46<5:09:15,  1.80s/it]
 30%|███       | 4445/14725 [2:17:47<5:09:07,  1.80s/it]
 30%|███       | 4446/14725 [2:17:49<5:11:23,  1.82s/it]
 30%|███       | 4447/14725 [2:17:51<5:09:07,  1.80s/it]
 30%|███       | 4448/14725 [2:17:53<5:10:55,  1.82s/it]
 30%|███       | 4449/14725 [2:17:55<5:10:36,  1.81s/it]
 30%|███       | 4450/14725 [2:17:56<5:09:13,  1.81s/it]
                                                        

 30%
+0: {'loss': 0.8558, 'grad_norm': 0.7341198707783336, 'learning_rate': 2e-05, 'epoch': 0.3}
+0: |███       | 4450/14725 [2:17:56<5:09:13,  1.81s/it]
 30%|███       | 4451/14725 [2:17:58<5:09:06,  1.81s/it]
 30%|███       | 4452/14725 [2:18:00<5:07:30,  1.80s/it]
 30%|███       | 4453/14725 [2:18:02<5:10:45,  1.82s/it]
 30%|███       | 4454/14725 [2:18:04<5:08:14,  1.80s/it]
 30%|███       | 4455/14725 [2:18:06<5:10:23,  1.81s/it]
 30%|███       | 4456/14725 [2:18:07<5:10:07,  1.81s/it]
 30%|███       | 4457/14725 [2:18:09<5:10:22,  1.81s/it]
 30%|███       | 4458/14725 [2:18:11<5:11:57,  1.82s/it]
 30%|███       | 4459/14725 [2:18:13<5:08:09,  1.80s/it]
 30%|███       | 4460/14725 [2:18:14<5:04:32,  1.78s/it]
                                                        

 30%|███       | 4460/14725 [2:18:14<5:04:32,  1.78s/it]
 30%|███       | 4461/14725 [2:18:16<5:05:00,  1.78s/it]
 30%|███       | 4462/14725 [2:18:18<5:09:41,  1.81s/it]
 30%|███       | 4463/14725 [2:18:20<5:09:34,  1.81s/it]
 30%|███       | 44
+0: {'loss': 0.8437, 'grad_norm': 0.7919698603403347, 'learning_rate': 2e-05, 'epoch': 0.3}
+0: 64/14725 [2:18:22<5:07:02,  1.80s/it]
 30%|███       | 4465/14725 [2:18:24<5:08:44,  1.81s/it]
 30%|███       | 4466/14725 [2:18:25<5:11:18,  1.82s/it]
 30%|███       | 4467/14725 [2:18:27<5:12:03,  1.83s/it]
 30%|███       | 4468/14725 [2:18:29<5:08:59,  1.81s/it]
 30%|███       | 4469/14725 [2:18:31<5:10:30,  1.82s/it]
 30%|███       | 4470/14725 [2:18:33<5:10:45,  1.82s/it]
                                                        

 30%|███       | 4470/14725 [2:18:33<5:10:45,  1.82s/it]
 30%|███       | 4471/14725 [2:18:34<5:08:53,  1.81s/it]
 30%|███       | 4472/14725 [2:18:36<5:06:26,  1.79s/it]
 30%|███       | 4473/14725 [2:18:38<5:07:30,  1.80s/it]
 30%|███       | 4474/14725 [2:18:40<5:08:29,  1.81s/it]
 30%|███       | 4475/14725 [2:18:42<5:05:39,  1.79s/it]
 30%|███       | 4476/14725 [2:18:43<5:08:14,  1.80s/it]
 30%|███       | 4477/14725 [2:18:45<5:05:59,  1.79s/it]
 30%|███       | 4478/14725 [2:18:47<5:0
+0: {'loss': 0.8447, 'grad_norm': 0.9286063599349158, 'learning_rate': 2e-05, 'epoch': 0.3}
+0: {'loss': 0.8351, 'grad_norm': 0.8216789751634516, 'learning_rate': 2e-05, 'epoch': 0.3}
+0: 9:09,  1.81s/it]
 30%|███       | 4479/14725 [2:18:49<5:06:34,  1.80s/it]
 30%|███       | 4480/14725 [2:18:51<5:08:25,  1.81s/it]
                                                        

 30%|███       | 4480/14725 [2:18:51<5:08:25,  1.81s/it]
 30%|███       | 4481/14725 [2:18:52<5:08:30,  1.81s/it]
 30%|███       | 4482/14725 [2:18:54<5:06:14,  1.79s/it]
 30%|███       | 4483/14725 [2:18:56<5:04:47,  1.79s/it]
 30%|███       | 4484/14725 [2:18:58<5:10:42,  1.82s/it]
 30%|███       | 4485/14725 [2:19:00<5:07:14,  1.80s/it]
 30%|███       | 4486/14725 [2:19:02<5:12:25,  1.83s/it]
 30%|███       | 4487/14725 [2:19:03<5:09:25,  1.81s/it]
 30%|███       | 4488/14725 [2:19:05<5:06:39,  1.80s/it]
 30%|███       | 4489/14725 [2:19:07<5:05:28,  1.79s/it]
 30%|███       | 4490/14725 [2:19:09<5:08:31,  1.81s/it]
                                                        

 30%|███       | 4490/14725 [2:19:09<5:08:31,  1.81s/it]
 30%|█�
+0: {'loss': 0.8621, 'grad_norm': 0.8408689489351123, 'learning_rate': 2e-05, 'epoch': 0.31}
+0: ��█       | 4491/14725 [2:19:10<5:06:53,  1.80s/it]
 31%|███       | 4492/14725 [2:19:12<5:13:06,  1.84s/it]
 31%|███       | 4493/14725 [2:19:14<5:13:22,  1.84s/it]
 31%|███       | 4494/14725 [2:19:16<5:13:43,  1.84s/it]
 31%|███       | 4495/14725 [2:19:18<5:10:07,  1.82s/it]
 31%|███       | 4496/14725 [2:19:20<5:10:09,  1.82s/it]
 31%|███       | 4497/14725 [2:19:21<5:11:32,  1.83s/it]
 31%|███       | 4498/14725 [2:19:23<5:14:42,  1.85s/it]
 31%|███       | 4499/14725 [2:19:25<5:10:14,  1.82s/it]
 31%|███       | 4500/14725 [2:19:27<5:10:48,  1.82s/it]
                                                        

 31%|███       | 4500/14725 [2:19:27<5:10:48,  1.82s/it]
 31%|███       | 4501/14725 [2:19:29<5:10:57,  1.82s/it]
 31%|███       | 4502/14725 [2:19:31<5:15:19,  1.85s/it]
 31%|███       | 4503/14725 [2:19:32<5:11:06,  1.83s/it]
 31%|███       | 4504/14725 [2:19:34<5:13:02,  1.84s/it]
 31%|███       | 4505/14
+0: {'loss': 0.8565, 'grad_norm': 0.7384343338863487, 'learning_rate': 2e-05, 'epoch': 0.31}
+0: 725 [2:19:36<5:09:40,  1.82s/it]
 31%|███       | 4506/14725 [2:19:38<5:10:59,  1.83s/it]
 31%|███       | 4507/14725 [2:19:40<5:13:35,  1.84s/it]
 31%|███       | 4508/14725 [2:19:42<5:15:13,  1.85s/it]
 31%|███       | 4509/14725 [2:19:44<5:14:05,  1.84s/it]
 31%|███       | 4510/14725 [2:19:45<5:11:06,  1.83s/it]
                                                        

 31%|███       | 4510/14725 [2:19:45<5:11:06,  1.83s/it]
 31%|███       | 4511/14725 [2:19:47<5:11:35,  1.83s/it]
 31%|███       | 4512/14725 [2:19:49<5:11:35,  1.83s/it]
 31%|███       | 4513/14725 [2:19:51<5:08:05,  1.81s/it]
 31%|███       | 4514/14725 [2:19:53<5:08:02,  1.81s/it]
 31%|███       | 4515/14725 [2:19:54<5:05:47,  1.80s/it]
 31%|███       | 4516/14725 [2:19:56<5:04:40,  1.79s/it]
 31%|███       | 4517/14725 [2:19:58<5:05:47,  1.80s/it]
 31%|███       | 4518/14725 [2:20:00<5:08:36,  1.81s/it]
 31%|███       | 4519/14725 [2:20:02<5:09:15,
+0: {'loss': 0.8241, 'grad_norm': 0.7248520223811149, 'learning_rate': 2e-05, 'epoch': 0.31}
+0: {'loss': 0.8511, 'grad_norm': 0.7555492358741646, 'learning_rate': 2e-05, 'epoch': 0.31}
+0:   1.82s/it]
 31%|███       | 4520/14725 [2:20:03<5:11:20,  1.83s/it]
                                                        

 31%|███       | 4520/14725 [2:20:03<5:11:20,  1.83s/it]
 31%|███       | 4521/14725 [2:20:05<5:10:04,  1.82s/it]
 31%|███       | 4522/14725 [2:20:07<5:07:18,  1.81s/it]
 31%|███       | 4523/14725 [2:20:09<5:06:02,  1.80s/it]
 31%|███       | 4524/14725 [2:20:11<5:05:52,  1.80s/it]
 31%|███       | 4525/14725 [2:20:12<5:04:58,  1.79s/it]
 31%|███       | 4526/14725 [2:20:14<5:09:32,  1.82s/it]
 31%|███       | 4527/14725 [2:20:16<5:13:11,  1.84s/it]
 31%|███       | 4528/14725 [2:20:18<5:10:12,  1.83s/it]
 31%|███       | 4529/14725 [2:20:20<5:07:52,  1.81s/it]
 31%|███       | 4530/14725 [2:20:22<5:06:45,  1.81s/it]
                                                        

 31%|███       | 4530/14725 [2:20:22<5:06:45,  1.81s/it]
 31%|███       | 4531/14725 [2:20:23<5:09:42,  1.82s/it]
 31%|███
+0: {'loss': 0.8406, 'grad_norm': 0.7690508115101257, 'learning_rate': 2e-05, 'epoch': 0.31}
+0:        | 4532/14725 [2:20:25<5:08:21,  1.82s/it]
 31%|███       | 4533/14725 [2:20:27<5:06:25,  1.80s/it]
 31%|███       | 4534/14725 [2:20:29<5:09:20,  1.82s/it]
 31%|███       | 4535/14725 [2:20:31<5:05:26,  1.80s/it]
 31%|███       | 4536/14725 [2:20:32<5:04:38,  1.79s/it]
 31%|███       | 4537/14725 [2:20:34<5:08:02,  1.81s/it]
 31%|███       | 4538/14725 [2:20:36<5:04:55,  1.80s/it]
 31%|███       | 4539/14725 [2:20:38<5:08:04,  1.81s/it]
 31%|███       | 4540/14725 [2:20:40<5:11:45,  1.84s/it]
                                                        

 31%|███       | 4540/14725 [2:20:40<5:11:45,  1.84s/it]
 31%|███       | 4541/14725 [2:20:42<5:09:42,  1.82s/it]
 31%|███       | 4542/14725 [2:20:43<5:10:34,  1.83s/it]
 31%|███       | 4543/14725 [2:20:45<5:08:29,  1.82s/it]
 31%|███       | 4544/14725 [2:20:47<5:08:26,  1.82s/it]
 31%|███       | 4545/14725 [2:20:49<5:08:16,  1.82s/it]
 31%|███       | 4546/14725 [
+0: {'loss': 0.837, 'grad_norm': 0.7252071817586212, 'learning_rate': 2e-05, 'epoch': 0.31}
+0: 2:20:51<5:05:27,  1.80s/it]
 31%|███       | 4547/14725 [2:20:52<5:04:07,  1.79s/it]
 31%|███       | 4548/14725 [2:20:54<5:04:41,  1.80s/it]
 31%|███       | 4549/14725 [2:20:56<5:06:09,  1.81s/it]
 31%|███       | 4550/14725 [2:20:58<5:05:19,  1.80s/it]
                                                        

 31%|███       | 4550/14725 [2:20:58<5:05:19,  1.80s/it]
 31%|███       | 4551/14725 [2:21:00<5:06:43,  1.81s/it]
 31%|███       | 4552/14725 [2:21:01<5:05:36,  1.80s/it]
 31%|███       | 4553/14725 [2:21:03<5:04:38,  1.80s/it]
 31%|███       | 4554/14725 [2:21:05<5:07:59,  1.82s/it]
 31%|███       | 4555/14725 [2:21:07<5:04:42,  1.80s/it]
 31%|███       | 4556/14725 [2:21:09<5:08:21,  1.82s/it]
 31%|███       | 4557/14725 [2:21:11<5:12:09,  1.84s/it]
 31%|███       | 4558/14725 [2:21:12<5:09:15,  1.83s/it]
 31%|███       | 4559/14725 [2:21:14<5:06:47,  1.81s/it]
 31%|███       | 4560/14725 [2:21:16<5:06:09,  1.8
+0: {'loss': 0.8474, 'grad_norm': 0.7661285913768359, 'learning_rate': 2e-05, 'epoch': 0.31}
+0: {'loss': 0.8568, 'grad_norm': 0.8072156266050613, 'learning_rate': 2e-05, 'epoch': 0.31}
+0: 1s/it]
                                                        

 31%|███       | 4560/14725 [2:21:16<5:06:09,  1.81s/it]
 31%|███       | 4561/14725 [2:21:18<5:06:48,  1.81s/it]
 31%|███       | 4562/14725 [2:21:20<5:07:23,  1.81s/it]
 31%|███       | 4563/14725 [2:21:21<5:09:24,  1.83s/it]
 31%|███       | 4564/14725 [2:21:23<5:07:50,  1.82s/it]
 31%|███       | 4565/14725 [2:21:25<5:05:28,  1.80s/it]
 31%|███       | 4566/14725 [2:21:27<5:05:50,  1.81s/it]
 31%|███       | 4567/14725 [2:21:29<5:04:41,  1.80s/it]
 31%|███       | 4568/14725 [2:21:30<5:06:29,  1.81s/it]
 31%|███       | 4569/14725 [2:21:32<5:05:51,  1.81s/it]
 31%|███       | 4570/14725 [2:21:34<5:05:31,  1.81s/it]
                                                        

 31%|███       | 4570/14725 [2:21:34<5:05:31,  1.81s/it]
 31%|███       | 4571/14725 [2:21:36<5:03:51,  1.80s/it]
 31%|███       | 4572/14725 [2:21:38<5:05:09,  1.80s/it]
 31%|███     
+0: {'loss': 0.8426, 'grad_norm': 0.7675145066242024, 'learning_rate': 2e-05, 'epoch': 0.31}
+0:   | 4573/14725 [2:21:39<5:04:01,  1.80s/it]
 31%|███       | 4574/14725 [2:21:41<5:01:37,  1.78s/it]
 31%|███       | 4575/14725 [2:21:43<5:00:13,  1.77s/it]
 31%|███       | 4576/14725 [2:21:45<4:59:47,  1.77s/it]
 31%|███       | 4577/14725 [2:21:46<5:01:24,  1.78s/it]
 31%|███       | 4578/14725 [2:21:48<5:01:35,  1.78s/it]
 31%|███       | 4579/14725 [2:21:50<5:06:32,  1.81s/it]
 31%|███       | 4580/14725 [2:21:52<5:07:37,  1.82s/it]
                                                        

 31%|███       | 4580/14725 [2:21:52<5:07:37,  1.82s/it]
 31%|███       | 4581/14725 [2:21:54<5:08:59,  1.83s/it]
 31%|███       | 4582/14725 [2:21:56<5:10:23,  1.84s/it]
 31%|███       | 4583/14725 [2:21:57<5:08:46,  1.83s/it]
 31%|███       | 4584/14725 [2:21:59<5:04:56,  1.80s/it]
 31%|███       | 4585/14725 [2:22:01<5:05:22,  1.81s/it]
 31%|███       | 4586/14725 [2:22:03<5:07:04,  1.82s/it]
 31%|███       | 4587/14725 [2:22:
+0: {'loss': 0.8344, 'grad_norm': 0.7469634855405766, 'learning_rate': 2e-05, 'epoch': 0.31}
+0: {'loss': 0.8318, 'grad_norm': 0.8476362801508869, 'learning_rate': 2e-05, 'epoch': 0.31}
+0: 05<5:03:47,  1.80s/it]
 31%|███       | 4588/14725 [2:22:06<5:05:39,  1.81s/it]
 31%|███       | 4589/14725 [2:22:08<5:07:05,  1.82s/it]
 31%|███       | 4590/14725 [2:22:10<5:05:30,  1.81s/it]
                                                        

 31%|███       | 4590/14725 [2:22:10<5:05:30,  1.81s/it]
 31%|███       | 4591/14725 [2:22:12<5:05:41,  1.81s/it]
 31%|███       | 4592/14725 [2:22:14<5:04:37,  1.80s/it]
 31%|███       | 4593/14725 [2:22:15<5:03:13,  1.80s/it]
 31%|███       | 4594/14725 [2:22:17<5:02:47,  1.79s/it]
 31%|███       | 4595/14725 [2:22:19<5:02:22,  1.79s/it]
 31%|███       | 4596/14725 [2:22:21<5:08:55,  1.83s/it]
 31%|███       | 4597/14725 [2:22:23<5:11:03,  1.84s/it]
 31%|███       | 4598/14725 [2:22:25<5:05:35,  1.81s/it]
 31%|███       | 4599/14725 [2:22:26<5:05:05,  1.81s/it]
 31%|███       | 4600/14725 [2:22:28<5:02:29,  1.79s/it]
                                                        

 31
+0: {'loss': 0.868, 'grad_norm': 0.7817928950846579, 'learning_rate': 2e-05, 'epoch': 0.31}
+0: %|███       | 4600/14725 [2:22:28<5:02:29,  1.79s/it]
 31%|███       | 4601/14725 [2:22:30<5:01:24,  1.79s/it]
 31%|███▏      | 4602/14725 [2:22:32<5:00:19,  1.78s/it]
 31%|███▏      | 4603/14725 [2:22:34<5:04:11,  1.80s/it]
 31%|███▏      | 4604/14725 [2:22:35<5:05:38,  1.81s/it]
 31%|███▏      | 4605/14725 [2:22:37<5:07:30,  1.82s/it]
 31%|███▏      | 4606/14725 [2:22:39<5:05:23,  1.81s/it]
 31%|███▏      | 4607/14725 [2:22:41<5:05:02,  1.81s/it]
 31%|███▏      | 4608/14725 [2:22:43<5:04:25,  1.81s/it]
 31%|███▏      | 4609/14725 [2:22:44<5:02:11,  1.79s/it]
 31%|███▏      | 4610/14725 [2:22:46<5:03:25,  1.80s/it]
                                                        

 31%|███▏      | 4610/14725 [2:22:46<5:03:25,  1.80s/it]
 31%|███▏      | 4611/14725 [2:22:48<5:02:31,  1.79s/it]
 31%|███▏      | 4612/14725 [2:22:50<5:00:43,  1.78s/it]
 31%|███▏      | 4613/14725 [2:22:52<5:03:20,  1.80s/it
+0: {'loss': 0.8404, 'grad_norm': 0.7859835421807158, 'learning_rate': 2e-05, 'epoch': 0.31}
+0: ]
 31%|███▏      | 4614/14725 [2:22:53<5:01:59,  1.79s/it]
 31%|███▏      | 4615/14725 [2:22:55<5:01:25,  1.79s/it]
 31%|███▏      | 4616/14725 [2:22:57<5:00:13,  1.78s/it]
 31%|███▏      | 4617/14725 [2:22:59<5:00:20,  1.78s/it]
 31%|███▏      | 4618/14725 [2:23:00<5:02:26,  1.80s/it]
 31%|███▏      | 4619/14725 [2:23:02<5:02:54,  1.80s/it]
 31%|███▏      | 4620/14725 [2:23:04<5:04:33,  1.81s/it]
                                                        

 31%|███▏      | 4620/14725 [2:23:04<5:04:33,  1.81s/it]
 31%|███▏      | 4621/14725 [2:23:06<5:06:05,  1.82s/it]
 31%|███▏      | 4622/14725 [2:23:08<5:04:22,  1.81s/it]
 31%|███▏      | 4623/14725 [2:23:10<5:08:32,  1.83s/it]
 31%|███▏      | 4624/14725 [2:23:11<5:07:48,  1.83s/it]
 31%|███▏      | 4625/14725 [2:23:13<5:05:02,  1.81s/it]
 31%|███▏      | 4626/14725 [2:23:15<5:06:49,  1.82s/it]
 31%|███▏      | 4627/14725 [2:23:17<5:04:20, 
+0: {'loss': 0.8574, 'grad_norm': 0.7634710463297834, 'learning_rate': 2e-05, 'epoch': 0.31}
+0:  1.81s/it]
 31%|███▏      | 4628/14725 [2:23:19<5:02:26,  1.80s/it]
 31%|███▏      | 4629/14725 [2:23:20<5:00:09,  1.78s/it]
 31%|███▏      | 4630/14725 [2:23:22<5:06:28,  1.82s/it]
                                                        

 31%|███▏      | 4630/14725 [2:23:22<5:06:28,  1.82s/it]
 31%|███▏      | 4631/14725 [2:23:24<5:03:55,  1.81s/it]
 31%|███▏      | 4632/14725 [2:23:26<5:02:35,  1.80s/it]
 31%|███▏      | 4633/14725 [2:23:28<5:03:03,  1.80s/it]
 31%|███▏      | 4634/14725 [2:23:29<5:02:19,  1.80s/it]
 31%|███▏      | 4635/14725 [2:23:31<5:04:36,  1.81s/it]
 31%|███▏      | 4636/14725 [2:23:33<5:05:37,  1.82s/it]
 31%|███▏      | 4637/14725 [2:23:35<5:03:02,  1.80s/it]
 31%|███▏      | 4638/14725 [2:23:37<5:01:13,  1.79s/it]
 32%|███▏      | 4639/14725 [2:23:38<5:02:06,  1.80s/it]
 32%|███▏      | 4640/14725 [2:23:40<5:05:57,  1.82s/it]
                                             
+0: {'loss': 0.8524, 'grad_norm': 0.753351637167829, 'learning_rate': 2e-05, 'epoch': 0.32}
+0: {'loss': 0.8352, 'grad_norm': 0.7708091479307997, 'learning_rate': 2e-05, 'epoch': 0.32}
+0:            

 32%|███▏      | 4640/14725 [2:23:40<5:05:57,  1.82s/it]
 32%|███▏      | 4641/14725 [2:23:42<5:08:43,  1.84s/it]
 32%|███▏      | 4642/14725 [2:23:44<5:05:56,  1.82s/it]
 32%|███▏      | 4643/14725 [2:23:46<5:03:27,  1.81s/it]
 32%|███▏      | 4644/14725 [2:23:48<5:04:50,  1.81s/it]
 32%|███▏      | 4645/14725 [2:23:49<5:05:51,  1.82s/it]
 32%|███▏      | 4646/14725 [2:23:51<5:07:20,  1.83s/it]
 32%|███▏      | 4647/14725 [2:23:53<5:08:33,  1.84s/it]
 32%|███▏      | 4648/14725 [2:23:55<5:03:42,  1.81s/it]
 32%|███▏      | 4649/14725 [2:23:57<5:02:00,  1.80s/it]
 32%|███▏      | 4650/14725 [2:23:58<5:01:32,  1.80s/it]
                                                        

 32%|███▏      | 4650/14725 [2:23:58<5:01:32,  1.80s/it]
 32%|███▏      | 4651/14725 [2:24:00<5:01:12,  1.79s/it]
 32%|███▏      | 4652/14725 [2:24:02<5:03:41,  1.81s/it]
 32%|███▏      | 4653/14725 [2:24:0
+0: {'loss': 0.844, 'grad_norm': 0.7500504386449215, 'learning_rate': 2e-05, 'epoch': 0.32}
+0: 4<5:02:06,  1.80s/it]
 32%|███▏      | 4654/14725 [2:24:06<5:01:45,  1.80s/it]
 32%|███▏      | 4655/14725 [2:24:07<5:02:09,  1.80s/it]
 32%|███▏      | 4656/14725 [2:24:09<5:00:43,  1.79s/it]
 32%|███▏      | 4657/14725 [2:24:11<5:05:59,  1.82s/it]
 32%|███▏      | 4658/14725 [2:24:13<5:05:29,  1.82s/it]
 32%|███▏      | 4659/14725 [2:24:15<5:05:33,  1.82s/it]
 32%|███▏      | 4660/14725 [2:24:17<5:03:52,  1.81s/it]
                                                        

 32%|███▏      | 4660/14725 [2:24:17<5:03:52,  1.81s/it]
 32%|███▏      | 4661/14725 [2:24:18<5:06:31,  1.83s/it]
 32%|███▏      | 4662/14725 [2:24:20<5:03:47,  1.81s/it]
 32%|███▏      | 4663/14725 [2:24:22<5:02:37,  1.80s/it]
 32%|███▏      | 4664/14725 [2:24:24<5:01:09,  1.80s/it]
 32%|███▏      | 4665/14725 [2:24:25<4:59:51,  1.79s/it]
 32%|███▏      | 4666/14725 [2:24:27<4:59:48,  1.79s/it]
 32%|███▏      | 4667/1472
+0: {'loss': 0.8279, 'grad_norm': 0.7806175048931718, 'learning_rate': 2e-05, 'epoch': 0.32}
+0: 5 [2:24:29<5:00:37,  1.79s/it]
 32%|███▏      | 4668/14725 [2:24:31<4:59:31,  1.79s/it]
 32%|███▏      | 4669/14725 [2:24:33<4:58:49,  1.78s/it]
 32%|███▏      | 4670/14725 [2:24:34<4:57:59,  1.78s/it]
                                                        

 32%|███▏      | 4670/14725 [2:24:34<4:57:59,  1.78s/it]
 32%|███▏      | 4671/14725 [2:24:36<5:02:29,  1.81s/it]
 32%|███▏      | 4672/14725 [2:24:38<5:00:24,  1.79s/it]
 32%|███▏      | 4673/14725 [2:24:40<5:05:09,  1.82s/it]
 32%|███▏      | 4674/14725 [2:24:42<5:05:12,  1.82s/it]
 32%|███▏      | 4675/14725 [2:24:44<5:03:41,  1.81s/it]
 32%|███▏      | 4676/14725 [2:24:45<5:02:38,  1.81s/it]
 32%|███▏      | 4677/14725 [2:24:47<5:01:01,  1.80s/it]
 32%|███▏      | 4678/14725 [2:24:49<5:01:28,  1.80s/it]
 32%|███▏      | 4679/14725 [2:24:51<5:02:01,  1.80s/it]
 32%|███▏      | 4680/14725 [2:24:53<5:07:07,  1.83s/it]
                         
+0: {'loss': 0.8498, 'grad_norm': 0.7039502004026544, 'learning_rate': 2e-05, 'epoch': 0.32}
+0: {'loss': 0.826, 'grad_norm': 0.7271338334126681, 'learning_rate': 2e-05, 'epoch': 0.32}
+0:                                

 32%|███▏      | 4680/14725 [2:24:53<5:07:07,  1.83s/it]
 32%|███▏      | 4681/14725 [2:24:54<5:04:35,  1.82s/it]
 32%|███▏      | 4682/14725 [2:24:56<5:04:00,  1.82s/it]
 32%|███▏      | 4683/14725 [2:24:58<5:02:35,  1.81s/it]
 32%|███▏      | 4684/14725 [2:25:00<5:00:49,  1.80s/it]
 32%|███▏      | 4685/14725 [2:25:02<5:02:56,  1.81s/it]
 32%|███▏      | 4686/14725 [2:25:03<5:01:30,  1.80s/it]
 32%|███▏      | 4687/14725 [2:25:05<5:03:46,  1.82s/it]
 32%|███▏      | 4688/14725 [2:25:07<5:05:35,  1.83s/it]
 32%|███▏      | 4689/14725 [2:25:09<5:05:02,  1.82s/it]
 32%|███▏      | 4690/14725 [2:25:11<5:03:17,  1.81s/it]
                                                        

 32%|███▏      | 4690/14725 [2:25:11<5:03:17,  1.81s/it]
 32%|███▏      | 4691/14725 [2:25:13<5:04:00,  1.82s/it]
 32%|███▏      | 4692/14725 [2:25:14<5:03:47,  1.82s/it]
 32%|███▏      
+0: {'loss': 0.862, 'grad_norm': 0.7654248887107106, 'learning_rate': 2e-05, 'epoch': 0.32}
+0: | 4693/14725 [2:25:16<5:04:01,  1.82s/it]
 32%|███▏      | 4694/14725 [2:25:18<5:00:26,  1.80s/it]
 32%|███▏      | 4695/14725 [2:25:20<5:11:07,  1.86s/it]
 32%|███▏      | 4696/14725 [2:25:22<5:07:56,  1.84s/it]
 32%|███▏      | 4697/14725 [2:25:24<5:08:19,  1.84s/it]
 32%|███▏      | 4698/14725 [2:25:25<5:04:21,  1.82s/it]
 32%|███▏      | 4699/14725 [2:25:27<5:03:04,  1.81s/it]
 32%|███▏      | 4700/14725 [2:25:29<5:07:29,  1.84s/it]
                                                        

 32%|███▏      | 4700/14725 [2:25:29<5:07:29,  1.84s/it]
 32%|███▏      | 4701/14725 [2:25:31<5:06:50,  1.84s/it]
 32%|███▏      | 4702/14725 [2:25:33<5:03:04,  1.81s/it]
 32%|███▏      | 4703/14725 [2:25:34<5:03:25,  1.82s/it]
 32%|███▏      | 4704/14725 [2:25:36<5:02:45,  1.81s/it]
 32%|███▏      | 4705/14725 [2:25:38<5:03:13,  1.82s/it]
 32%|███▏      | 4706/14725 [2:25:40<5:01:17,  1.80s/it]
 32%|███
+0: {'loss': 0.8353, 'grad_norm': 0.7921818994078429, 'learning_rate': 2e-05, 'epoch': 0.32}
+0: ▏      | 4707/14725 [2:25:42<5:01:44,  1.81s/it]
 32%|███▏      | 4708/14725 [2:25:43<5:01:25,  1.81s/it]
 32%|███▏      | 4709/14725 [2:25:45<5:00:42,  1.80s/it]
 32%|███▏      | 4710/14725 [2:25:47<5:04:19,  1.82s/it]
                                                        

 32%|███▏      | 4710/14725 [2:25:47<5:04:19,  1.82s/it]
 32%|███▏      | 4711/14725 [2:25:49<5:01:11,  1.80s/it]
 32%|███▏      | 4712/14725 [2:25:51<5:10:45,  1.86s/it]
 32%|███▏      | 4713/14725 [2:25:53<5:09:24,  1.85s/it]
 32%|███▏      | 4714/14725 [2:25:55<5:08:24,  1.85s/it]
 32%|███▏      | 4715/14725 [2:25:56<5:03:19,  1.82s/it]
 32%|███▏      | 4716/14725 [2:25:58<5:01:32,  1.81s/it]
 32%|███▏      | 4717/14725 [2:26:00<5:01:58,  1.81s/it]
 32%|███▏      | 4718/14725 [2:26:02<5:05:48,  1.83s/it]
 32%|███▏      | 4719/14725 [2:26:04<5:02:22,  1.81s/it]
 32%|███▏      | 4720/14725 [2:26:05<5:03:24,  1.82s/it]
     
+0: {'loss': 0.8385, 'grad_norm': 0.8015010746502752, 'learning_rate': 2e-05, 'epoch': 0.32}
+0: {'loss': 0.8564, 'grad_norm': 0.8330449935072302, 'learning_rate': 2e-05, 'epoch': 0.32}
+0:                                                    

 32%|███▏      | 4720/14725 [2:26:05<5:03:24,  1.82s/it]
 32%|███▏      | 4721/14725 [2:26:07<5:08:12,  1.85s/it]
 32%|███▏      | 4722/14725 [2:26:09<5:06:34,  1.84s/it]
 32%|███▏      | 4723/14725 [2:26:11<5:04:32,  1.83s/it]
 32%|███▏      | 4724/14725 [2:26:13<5:00:36,  1.80s/it]
 32%|███▏      | 4725/14725 [2:26:15<5:03:36,  1.82s/it]
 32%|███▏      | 4726/14725 [2:26:16<5:01:23,  1.81s/it]
 32%|███▏      | 4727/14725 [2:26:18<5:00:40,  1.80s/it]
 32%|███▏      | 4728/14725 [2:26:20<5:02:21,  1.81s/it]
 32%|███▏      | 4729/14725 [2:26:22<5:03:14,  1.82s/it]
 32%|███▏      | 4730/14725 [2:26:24<5:02:47,  1.82s/it]
                                                        

 32%|███▏      | 4730/14725 [2:26:24<5:02:47,  1.82s/it]
 32%|███▏      | 4731/14725 [2:26:25<5:04:54,  1.83s/it]
 32%|███▏      | 4732/14725 [2:26:27<5:03:38,  1.82s/it]
 32
+0: {'loss': 0.8435, 'grad_norm': 0.7978080450100749, 'learning_rate': 2e-05, 'epoch': 0.32}
+0: %|███▏      | 4733/14725 [2:26:29<5:03:11,  1.82s/it]
 32%|███▏      | 4734/14725 [2:26:31<5:06:46,  1.84s/it]
 32%|███▏      | 4735/14725 [2:26:33<5:02:02,  1.81s/it]
 32%|███▏      | 4736/14725 [2:26:35<5:02:56,  1.82s/it]
 32%|███▏      | 4737/14725 [2:26:36<5:02:28,  1.82s/it]
 32%|███▏      | 4738/14725 [2:26:38<4:59:19,  1.80s/it]
 32%|███▏      | 4739/14725 [2:26:40<4:57:19,  1.79s/it]
 32%|███▏      | 4740/14725 [2:26:42<5:01:24,  1.81s/it]
                                                        

 32%|███▏      | 4740/14725 [2:26:42<5:01:24,  1.81s/it]
 32%|███▏      | 4741/14725 [2:26:44<5:03:42,  1.83s/it]
 32%|███▏      | 4742/14725 [2:26:45<5:03:41,  1.83s/it]
 32%|███▏      | 4743/14725 [2:26:47<4:59:50,  1.80s/it]
 32%|███▏      | 4744/14725 [2:26:49<4:58:47,  1.80s/it]
 32%|███▏      | 4745/14725 [2:26:51<4:59:12,  1.80s/it]
 32%|███▏      | 4746/14725 [2:26:53<5:03:56,  1.83
+0: {'loss': 0.86, 'grad_norm': 0.7880882003834201, 'learning_rate': 2e-05, 'epoch': 0.32}
+0: s/it]
 32%|███▏      | 4747/14725 [2:26:55<5:09:03,  1.86s/it]
 32%|███▏      | 4748/14725 [2:26:56<5:08:02,  1.85s/it]
 32%|███▏      | 4749/14725 [2:26:58<5:02:47,  1.82s/it]
 32%|███▏      | 4750/14725 [2:27:00<5:01:13,  1.81s/it]
                                                        

 32%|███▏      | 4750/14725 [2:27:00<5:01:13,  1.81s/it]
 32%|███▏      | 4751/14725 [2:27:02<5:00:49,  1.81s/it]
 32%|███▏      | 4752/14725 [2:27:04<5:01:43,  1.82s/it]
 32%|███▏      | 4753/14725 [2:27:05<5:01:01,  1.81s/it]
 32%|███▏      | 4754/14725 [2:27:07<5:00:15,  1.81s/it]
 32%|███▏      | 4755/14725 [2:27:09<4:57:17,  1.79s/it]
 32%|███▏      | 4756/14725 [2:27:11<4:57:01,  1.79s/it]
 32%|███▏      | 4757/14725 [2:27:13<4:59:18,  1.80s/it]
 32%|███▏      | 4758/14725 [2:27:14<4:58:35,  1.80s/it]
 32%|███▏      | 4759/14725 [2:27:16<4:55:41,  1.78s/it]
 32%|███▏      | 4760/14725 [2:27:18<4:55:
+0: {'loss': 0.8198, 'grad_norm': 0.8760543939136619, 'learning_rate': 2e-05, 'epoch': 0.32}
+0: {'loss': 0.8445, 'grad_norm': 0.730208951431563, 'learning_rate': 2e-05, 'epoch': 0.32}
+0: 34,  1.78s/it]
                                                        

 32%|███▏      | 4760/14725 [2:27:18<4:55:34,  1.78s/it]
 32%|███▏      | 4761/14725 [2:27:20<4:55:11,  1.78s/it]
 32%|███▏      | 4762/14725 [2:27:21<4:57:27,  1.79s/it]
 32%|███▏      | 4763/14725 [2:27:23<4:54:39,  1.77s/it]
 32%|███▏      | 4764/14725 [2:27:25<4:54:59,  1.78s/it]
 32%|███▏      | 4765/14725 [2:27:27<4:59:27,  1.80s/it]
 32%|███▏      | 4766/14725 [2:27:29<4:59:58,  1.81s/it]
 32%|███▏      | 4767/14725 [2:27:30<4:59:20,  1.80s/it]
 32%|███▏      | 4768/14725 [2:27:32<4:56:28,  1.79s/it]
 32%|███▏      | 4769/14725 [2:27:34<4:55:54,  1.78s/it]
 32%|███▏      | 4770/14725 [2:27:36<4:58:20,  1.80s/it]
                                                        

 32%|███▏      | 4770/14725 [2:27:36<4:58:20,  1.80s/it]
 32%|███▏      | 4771/14725 [2:27:38<4:57:34,  1.79s/it]
 32%|███▏      | 4772/14725 [2:27:39<4:5
+0: {'loss': 0.8501, 'grad_norm': 0.7873689386830812, 'learning_rate': 2e-05, 'epoch': 0.32}
+0: 4:55,  1.78s/it]
 32%|███▏      | 4773/14725 [2:27:41<4:55:44,  1.78s/it]
 32%|███▏      | 4774/14725 [2:27:43<5:03:27,  1.83s/it]
 32%|███▏      | 4775/14725 [2:27:45<5:08:20,  1.86s/it]
 32%|███▏      | 4776/14725 [2:27:47<5:08:32,  1.86s/it]
 32%|███▏      | 4777/14725 [2:27:49<5:06:33,  1.85s/it]
 32%|███▏      | 4778/14725 [2:27:50<5:02:12,  1.82s/it]
 32%|███▏      | 4779/14725 [2:27:52<5:11:29,  1.88s/it]
 32%|███▏      | 4780/14725 [2:27:54<5:08:45,  1.86s/it]
                                                        

 32%|███▏      | 4780/14725 [2:27:54<5:08:45,  1.86s/it]
 32%|███▏      | 4781/14725 [2:27:56<5:11:23,  1.88s/it]
 32%|███▏      | 4782/14725 [2:27:58<5:08:24,  1.86s/it]
 32%|███▏      | 4783/14725 [2:28:00<5:04:55,  1.84s/it]
 32%|███▏      | 4784/14725 [2:28:02<5:01:29,  1.82s/it]
 32%|███▏      | 4785/14725 [2:28:03<4:57:19,  1.79s/it]
 33%|███▎      | 4786/14725 [2:
+0: {'loss': 0.8382, 'grad_norm': 0.7294917654682925, 'learning_rate': 2e-05, 'epoch': 0.33}
+0: 28:05<4:56:19,  1.79s/it]
 33%|███▎      | 4787/14725 [2:28:07<4:58:27,  1.80s/it]
 33%|███▎      | 4788/14725 [2:28:09<4:58:31,  1.80s/it]
 33%|███▎      | 4789/14725 [2:28:11<4:56:10,  1.79s/it]
 33%|███▎      | 4790/14725 [2:28:12<4:54:49,  1.78s/it]
                                                        

 33%|███▎      | 4790/14725 [2:28:12<4:54:49,  1.78s/it]
 33%|███▎      | 4791/14725 [2:28:14<4:57:13,  1.80s/it]
 33%|███▎      | 4792/14725 [2:28:16<4:59:10,  1.81s/it]
 33%|███▎      | 4793/14725 [2:28:18<4:58:02,  1.80s/it]
 33%|███▎      | 4794/14725 [2:28:20<4:59:50,  1.81s/it]
 33%|███▎      | 4795/14725 [2:28:21<4:59:42,  1.81s/it]
 33%|███▎      | 4796/14725 [2:28:23<4:58:08,  1.80s/it]
 33%|███▎      | 4797/14725 [2:28:25<5:00:24,  1.82s/it]
 33%|███▎      | 4798/14725 [2:28:27<4:56:55,  1.79s/it]
 33%|███▎      | 4799/14725 [2:28:29<5:00:05,  1.81s/it]
 33%|███▎      | 4800/
+0: {'loss': 0.845, 'grad_norm': 0.7716646935345323, 'learning_rate': 2e-05, 'epoch': 0.33}
+0: {'loss': 0.8368, 'grad_norm': 0.7365335731085277, 'learning_rate': 2e-05, 'epoch': 0.33}
+0: 14725 [2:28:30<4:56:20,  1.79s/it]
                                                        

 33%|███▎      | 4800/14725 [2:28:30<4:56:20,  1.79s/it]
 33%|███▎      | 4801/14725 [2:28:32<4:57:16,  1.80s/it]
 33%|███▎      | 4802/14725 [2:28:34<4:57:23,  1.80s/it]
 33%|███▎      | 4803/14725 [2:28:36<4:57:15,  1.80s/it]
 33%|███▎      | 4804/14725 [2:28:38<4:57:13,  1.80s/it]
 33%|███▎      | 4805/14725 [2:28:39<4:54:53,  1.78s/it]
 33%|███▎      | 4806/14725 [2:28:41<4:57:14,  1.80s/it]
 33%|███▎      | 4807/14725 [2:28:43<4:56:35,  1.79s/it]
 33%|███▎      | 4808/14725 [2:28:45<4:54:11,  1.78s/it]
 33%|███▎      | 4809/14725 [2:28:47<4:58:28,  1.81s/it]
 33%|███▎      | 4810/14725 [2:28:48<4:57:15,  1.80s/it]
                                                        

 33%|███▎      | 4810/14725 [2:28:48<4:57:15,  1.80s/it]
 33%|███▎      | 4811/14725 [2:28:50<5:01:24,  1.82s/it]
 33%|███▎      | 481
+0: {'loss': 0.8635, 'grad_norm': 0.7681018026368481, 'learning_rate': 2e-05, 'epoch': 0.33}
+0: 2/14725 [2:28:52<5:03:03,  1.83s/it]
 33%|███▎      | 4813/14725 [2:28:54<5:03:46,  1.84s/it]
 33%|███▎      | 4814/14725 [2:28:56<5:01:10,  1.82s/it]
 33%|███▎      | 4815/14725 [2:28:58<5:06:31,  1.86s/it]
 33%|███▎      | 4816/14725 [2:28:59<5:02:08,  1.83s/it]
 33%|███▎      | 4817/14725 [2:29:01<5:02:30,  1.83s/it]
 33%|███▎      | 4818/14725 [2:29:03<5:00:43,  1.82s/it]
 33%|███▎      | 4819/14725 [2:29:05<4:58:01,  1.81s/it]
 33%|███▎      | 4820/14725 [2:29:07<4:56:39,  1.80s/it]
                                                        

 33%|███▎      | 4820/14725 [2:29:07<4:56:39,  1.80s/it]
 33%|███▎      | 4821/14725 [2:29:08<4:59:44,  1.82s/it]
 33%|███▎      | 4822/14725 [2:29:10<4:59:28,  1.81s/it]
 33%|███▎      | 4823/14725 [2:29:12<4:57:34,  1.80s/it]
 33%|███▎      | 4824/14725 [2:29:14<4:59:12,  1.81s/it]
 33%|███▎      | 4825/14725 [2:29:16<4:58:00,  1.81s/it]
 33%|███▎  
+0: {'loss': 0.8403, 'grad_norm': 0.6847389149999454, 'learning_rate': 2e-05, 'epoch': 0.33}
+0:     | 4826/14725 [2:29:17<4:58:35,  1.81s/it]
 33%|███▎      | 4827/14725 [2:29:19<4:59:51,  1.82s/it]
 33%|███▎      | 4828/14725 [2:29:21<4:58:36,  1.81s/it]
 33%|███▎      | 4829/14725 [2:29:23<4:58:15,  1.81s/it]
 33%|███▎      | 4830/14725 [2:29:25<4:59:48,  1.82s/it]
                                                        

 33%|███▎      | 4830/14725 [2:29:25<4:59:48,  1.82s/it]
 33%|███▎      | 4831/14725 [2:29:27<5:02:49,  1.84s/it]
 33%|███▎      | 4832/14725 [2:29:28<5:01:20,  1.83s/it]
 33%|███▎      | 4833/14725 [2:29:30<4:57:39,  1.81s/it]
 33%|███▎      | 4834/14725 [2:29:32<5:00:51,  1.83s/it]
 33%|███▎      | 4835/14725 [2:29:34<5:02:30,  1.84s/it]
 33%|███▎      | 4836/14725 [2:29:36<5:01:35,  1.83s/it]
 33%|███▎      | 4837/14725 [2:29:37<4:57:38,  1.81s/it]
 33%|███▎      | 4838/14725 [2:29:39<4:57:11,  1.80s/it]
 33%|███▎      | 4839/14725 [2:29:41<4:58:45,  1.81s/it]
 33%|█�
+0: {'loss': 0.8453, 'grad_norm': 0.780574278982286, 'learning_rate': 2e-05, 'epoch': 0.33}
+0: {'loss': 0.8575, 'grad_norm': 0.7488210043765745, 'learning_rate': 2e-05, 'epoch': 0.33}
+0: �█▎      | 4840/14725 [2:29:43<4:56:03,  1.80s/it]
                                                        

 33%|███▎      | 4840/14725 [2:29:43<4:56:03,  1.80s/it]
 33%|███▎      | 4841/14725 [2:29:45<4:58:38,  1.81s/it]
 33%|███▎      | 4842/14725 [2:29:46<4:57:23,  1.81s/it]
 33%|███▎      | 4843/14725 [2:29:48<4:55:16,  1.79s/it]
 33%|███▎      | 4844/14725 [2:29:50<4:58:35,  1.81s/it]
 33%|███▎      | 4845/14725 [2:29:52<4:57:41,  1.81s/it]
 33%|███▎      | 4846/14725 [2:29:54<4:58:02,  1.81s/it]
 33%|███▎      | 4847/14725 [2:29:56<5:00:29,  1.83s/it]
 33%|███▎      | 4848/14725 [2:29:57<4:59:30,  1.82s/it]
 33%|███▎      | 4849/14725 [2:29:59<4:57:14,  1.81s/it]
 33%|███▎      | 4850/14725 [2:30:01<4:58:24,  1.81s/it]
                                                        

 33%|███▎      | 4850/14725 [2:30:01<4:58:24,  1.81s/it]
 33%|███▎      | 4851/14725 [2:30:03<4:59:34,  1.82s/it]
 33%|█
+0: {'loss': 0.8637, 'grad_norm': 0.7707905000899502, 'learning_rate': 2e-05, 'epoch': 0.33}
+0: ██▎      | 4852/14725 [2:30:05<4:57:38,  1.81s/it]
 33%|███▎      | 4853/14725 [2:30:06<4:58:52,  1.82s/it]
 33%|███▎      | 4854/14725 [2:30:08<4:55:34,  1.80s/it]
 33%|███▎      | 4855/14725 [2:30:10<4:57:54,  1.81s/it]
 33%|███▎      | 4856/14725 [2:30:12<4:59:31,  1.82s/it]
 33%|███▎      | 4857/14725 [2:30:14<4:57:45,  1.81s/it]
 33%|███▎      | 4858/14725 [2:30:16<5:00:23,  1.83s/it]
 33%|███▎      | 4859/14725 [2:30:17<4:59:47,  1.82s/it]
 33%|███▎      | 4860/14725 [2:30:19<4:58:23,  1.81s/it]
                                                        

 33%|███▎      | 4860/14725 [2:30:19<4:58:23,  1.81s/it]
 33%|███▎      | 4861/14725 [2:30:21<5:01:25,  1.83s/it]
 33%|███▎      | 4862/14725 [2:30:23<4:58:32,  1.82s/it]
 33%|███▎      | 4863/14725 [2:30:25<4:54:57,  1.79s/it]
 33%|███▎      | 4864/14725 [2:30:26<4:54:22,  1.79s/it]
 33%|███▎      | 4865/14725 [2:30:28<4:53:50,  1.79s/it]
+0: {'loss': 0.8488, 'grad_norm': 0.753594882917903, 'learning_rate': 2e-05, 'epoch': 0.33}
+0: 
 33%|███▎      | 4866/14725 [2:30:30<4:55:07,  1.80s/it]
 33%|███▎      | 4867/14725 [2:30:32<4:56:48,  1.81s/it]
 33%|███▎      | 4868/14725 [2:30:34<4:59:12,  1.82s/it]
 33%|███▎      | 4869/14725 [2:30:35<5:00:15,  1.83s/it]
 33%|███▎      | 4870/14725 [2:30:37<4:59:08,  1.82s/it]
                                                        

 33%|███▎      | 4870/14725 [2:30:37<4:59:08,  1.82s/it]
 33%|███▎      | 4871/14725 [2:30:39<4:58:38,  1.82s/it]
 33%|███▎      | 4872/14725 [2:30:41<5:00:02,  1.83s/it]
 33%|███▎      | 4873/14725 [2:30:43<4:59:50,  1.83s/it]
 33%|███▎      | 4874/14725 [2:30:45<4:58:44,  1.82s/it]
 33%|███▎      | 4875/14725 [2:30:46<5:02:44,  1.84s/it]
 33%|███▎      | 4876/14725 [2:30:48<4:59:46,  1.83s/it]
 33%|███▎      | 4877/14725 [2:30:50<4:55:15,  1.80s/it]
 33%|███▎      | 4878/14725 [2:30:52<4:59:17,  1.82s/it]
 33%|███▎      | 4879/14725 [2:30:54<4:55:45,  
+0: {'loss': 0.8526, 'grad_norm': 0.7316963882889166, 'learning_rate': 2e-05, 'epoch': 0.33}
+0: {'loss': 0.8511, 'grad_norm': 0.798430565393836, 'learning_rate': 2e-05, 'epoch': 0.33}
+0: 1.80s/it]
 33%|███▎      | 4880/14725 [2:30:56<5:00:11,  1.83s/it]
                                                        

 33%|███▎      | 4880/14725 [2:30:56<5:00:11,  1.83s/it]
 33%|███▎      | 4881/14725 [2:30:57<5:00:47,  1.83s/it]
 33%|███▎      | 4882/14725 [2:30:59<5:00:11,  1.83s/it]
 33%|███▎      | 4883/14725 [2:31:01<4:56:27,  1.81s/it]
 33%|███▎      | 4884/14725 [2:31:03<4:57:03,  1.81s/it]
 33%|███▎      | 4885/14725 [2:31:04<4:53:28,  1.79s/it]
 33%|███▎      | 4886/14725 [2:31:06<4:56:35,  1.81s/it]
 33%|███▎      | 4887/14725 [2:31:08<4:54:35,  1.80s/it]
 33%|███▎      | 4888/14725 [2:31:10<5:03:27,  1.85s/it]
 33%|███▎      | 4889/14725 [2:31:12<5:00:33,  1.83s/it]
 33%|███▎      | 4890/14725 [2:31:14<5:03:15,  1.85s/it]
                                                        

 33%|███▎      | 4890/14725 [2:31:14<5:03:15,  1.85s/it]
 33%|███▎      | 4891/14725 [2:31:16<5:02:15,
+0: {'loss': 0.834, 'grad_norm': 0.7968949234536974, 'learning_rate': 2e-05, 'epoch': 0.33}
+0:   1.84s/it]
 33%|███▎      | 4892/14725 [2:31:17<4:59:55,  1.83s/it]
 33%|███▎      | 4893/14725 [2:31:19<5:00:10,  1.83s/it]
 33%|███▎      | 4894/14725 [2:31:21<4:57:16,  1.81s/it]
 33%|███▎      | 4895/14725 [2:31:23<4:55:40,  1.80s/it]
 33%|███▎      | 4896/14725 [2:31:25<4:53:01,  1.79s/it]
 33%|███▎      | 4897/14725 [2:31:26<4:55:30,  1.80s/it]
 33%|███▎      | 4898/14725 [2:31:28<4:52:29,  1.79s/it]
 33%|███▎      | 4899/14725 [2:31:30<4:51:30,  1.78s/it]
 33%|███▎      | 4900/14725 [2:31:32<4:51:56,  1.78s/it]
                                                        

 33%|███▎      | 4900/14725 [2:31:32<4:51:56,  1.78s/it]
 33%|███▎      | 4901/14725 [2:31:33<4:52:12,  1.78s/it]
 33%|███▎      | 4902/14725 [2:31:35<4:51:43,  1.78s/it]
 33%|███▎      | 4903/14725 [2:31:37<4:55:42,  1.81s/it]
 33%|███▎      | 4904/14725 [2:31:39<4:57:02,  1.81s/it]
 33%|███▎      | 4905/14725 [2:31:41
+0: {'loss': 0.8424, 'grad_norm': 0.7664751337388898, 'learning_rate': 2e-05, 'epoch': 0.33}
+0: <4:58:39,  1.82s/it]
 33%|███▎      | 4906/14725 [2:31:43<4:54:48,  1.80s/it]
 33%|███▎      | 4907/14725 [2:31:44<4:52:14,  1.79s/it]
 33%|███▎      | 4908/14725 [2:31:46<4:54:21,  1.80s/it]
 33%|███▎      | 4909/14725 [2:31:48<4:52:54,  1.79s/it]
 33%|███▎      | 4910/14725 [2:31:50<4:53:14,  1.79s/it]
                                                        

 33%|███▎      | 4910/14725 [2:31:50<4:53:14,  1.79s/it]
 33%|███▎      | 4911/14725 [2:31:52<4:55:56,  1.81s/it]
 33%|███▎      | 4912/14725 [2:31:53<4:53:57,  1.80s/it]
 33%|███▎      | 4913/14725 [2:31:55<4:55:48,  1.81s/it]
 33%|███▎      | 4914/14725 [2:31:57<4:54:00,  1.80s/it]
 33%|███▎      | 4915/14725 [2:31:59<4:53:20,  1.79s/it]
 33%|███▎      | 4916/14725 [2:32:00<4:52:18,  1.79s/it]
 33%|███▎      | 4917/14725 [2:32:02<4:54:16,  1.80s/it]
 33%|███▎      | 4918/14725 [2:32:04<4:52:54,  1.79s/it]
 33%|███▎      | 4919/14725
+0: {'loss': 0.8372, 'grad_norm': 0.7773435212480265, 'learning_rate': 2e-05, 'epoch': 0.33}
+0: {'loss': 0.8644, 'grad_norm': 0.7515017238777096, 'learning_rate': 2e-05, 'epoch': 0.33}
+0:  [2:32:06<4:53:02,  1.79s/it]
 33%|███▎      | 4920/14725 [2:32:08<4:52:06,  1.79s/it]
                                                        

 33%|███▎      | 4920/14725 [2:32:08<4:52:06,  1.79s/it]
 33%|███▎      | 4921/14725 [2:32:09<4:51:50,  1.79s/it]
 33%|███▎      | 4922/14725 [2:32:11<4:51:31,  1.78s/it]
 33%|███▎      | 4923/14725 [2:32:13<4:51:50,  1.79s/it]
 33%|███▎      | 4924/14725 [2:32:15<4:49:31,  1.77s/it]
 33%|███▎      | 4925/14725 [2:32:17<4:53:00,  1.79s/it]
 33%|███▎      | 4926/14725 [2:32:18<4:54:54,  1.81s/it]
 33%|███▎      | 4927/14725 [2:32:20<4:53:12,  1.80s/it]
 33%|███▎      | 4928/14725 [2:32:22<4:57:06,  1.82s/it]
 33%|███▎      | 4929/14725 [2:32:24<4:57:54,  1.82s/it]
 33%|███▎      | 4930/14725 [2:32:26<4:53:37,  1.80s/it]
                                                        

 33%|███▎      | 4930/14725 [2:32:26<4:53:37,  1.80s/it]
 33%|███▎      | 4931/147
+0: {'loss': 0.8496, 'grad_norm': 0.7641282963236005, 'learning_rate': 2e-05, 'epoch': 0.34}
+0: 25 [2:32:27<4:54:49,  1.81s/it]
 33%|███▎      | 4932/14725 [2:32:29<4:53:55,  1.80s/it]
 34%|███▎      | 4933/14725 [2:32:31<4:52:46,  1.79s/it]
 34%|███▎      | 4934/14725 [2:32:33<4:54:42,  1.81s/it]
 34%|███▎      | 4935/14725 [2:32:35<4:55:00,  1.81s/it]
 34%|███▎      | 4936/14725 [2:32:36<4:51:24,  1.79s/it]
 34%|███▎      | 4937/14725 [2:32:38<4:51:30,  1.79s/it]
 34%|███▎      | 4938/14725 [2:32:40<4:52:12,  1.79s/it]
 34%|███▎      | 4939/14725 [2:32:42<4:51:57,  1.79s/it]
 34%|███▎      | 4940/14725 [2:32:44<4:50:14,  1.78s/it]
                                                        

 34%|███▎      | 4940/14725 [2:32:44<4:50:14,  1.78s/it]
 34%|███▎      | 4941/14725 [2:32:45<4:49:59,  1.78s/it]
 34%|███▎      | 4942/14725 [2:32:47<4:50:47,  1.78s/it]
 34%|███▎      | 4943/14725 [2:32:49<4:50:10,  1.78s/it]
 34%|███▎      | 4944/14725 [2:32:51<4:48:41,  1.77s/it]
 34%|███▎      |
+0: {'loss': 0.8289, 'grad_norm': 0.714853053599359, 'learning_rate': 2e-05, 'epoch': 0.34}
+0:  4945/14725 [2:32:52<4:53:48,  1.80s/it]
 34%|███▎      | 4946/14725 [2:32:54<4:51:32,  1.79s/it]
 34%|███▎      | 4947/14725 [2:32:56<4:50:57,  1.79s/it]
 34%|███▎      | 4948/14725 [2:32:58<4:49:52,  1.78s/it]
 34%|███▎      | 4949/14725 [2:33:00<4:52:55,  1.80s/it]
 34%|███▎      | 4950/14725 [2:33:01<4:54:34,  1.81s/it]
                                                        

 34%|███▎      | 4950/14725 [2:33:01<4:54:34,  1.81s/it]
 34%|███▎      | 4951/14725 [2:33:03<4:58:51,  1.83s/it]
 34%|███▎      | 4952/14725 [2:33:05<5:00:08,  1.84s/it]
 34%|███▎      | 4953/14725 [2:33:07<4:58:45,  1.83s/it]
 34%|███▎      | 4954/14725 [2:33:09<5:01:20,  1.85s/it]
 34%|███▎      | 4955/14725 [2:33:11<4:59:30,  1.84s/it]
 34%|███▎      | 4956/14725 [2:33:13<4:57:18,  1.83s/it]
 34%|███▎      | 4957/14725 [2:33:14<4:54:38,  1.81s/it]
 34%|███▎      | 4958/14725 [2:33:16<4:54:19,  1.81s/it]
 34%|███�
+0: {'loss': 0.8458, 'grad_norm': 0.759820751700324, 'learning_rate': 2e-05, 'epoch': 0.34}
+0: {'loss': 0.8469, 'grad_norm': 0.7963023673669914, 'learning_rate': 2e-05, 'epoch': 0.34}
+0: ��      | 4959/14725 [2:33:18<4:54:34,  1.81s/it]
 34%|███▎      | 4960/14725 [2:33:20<4:51:59,  1.79s/it]
                                                        

 34%|███▎      | 4960/14725 [2:33:20<4:51:59,  1.79s/it]
 34%|███▎      | 4961/14725 [2:33:22<4:54:20,  1.81s/it]
 34%|███▎      | 4962/14725 [2:33:23<4:53:24,  1.80s/it]
 34%|███▎      | 4963/14725 [2:33:25<4:51:03,  1.79s/it]
 34%|███▎      | 4964/14725 [2:33:27<4:52:51,  1.80s/it]
 34%|███▎      | 4965/14725 [2:33:29<4:55:26,  1.82s/it]
 34%|███▎      | 4966/14725 [2:33:31<4:56:46,  1.82s/it]
 34%|███▎      | 4967/14725 [2:33:32<4:55:46,  1.82s/it]
 34%|███▎      | 4968/14725 [2:33:34<4:53:40,  1.81s/it]
 34%|███▎      | 4969/14725 [2:33:36<4:58:16,  1.83s/it]
 34%|███▍      | 4970/14725 [2:33:38<4:59:06,  1.84s/it]
                                                        

 34%|███▍      | 4970/14725 [2:33:38<4:59:06,  1.84s/it]
 34%|██�
+0: {'loss': 0.8334, 'grad_norm': 0.7829467568822742, 'learning_rate': 2e-05, 'epoch': 0.34}
+0: �▍      | 4971/14725 [2:33:40<4:57:25,  1.83s/it]
 34%|███▍      | 4972/14725 [2:33:42<5:01:50,  1.86s/it]
 34%|███▍      | 4973/14725 [2:33:43<5:00:16,  1.85s/it]
 34%|███▍      | 4974/14725 [2:33:45<4:59:59,  1.85s/it]
 34%|███▍      | 4975/14725 [2:33:47<4:59:20,  1.84s/it]
 34%|███▍      | 4976/14725 [2:33:49<4:57:22,  1.83s/it]
 34%|███▍      | 4977/14725 [2:33:51<4:57:39,  1.83s/it]
 34%|███▍      | 4978/14725 [2:33:53<4:55:32,  1.82s/it]
 34%|███▍      | 4979/14725 [2:33:54<4:57:50,  1.83s/it]
 34%|███▍      | 4980/14725 [2:33:56<4:59:16,  1.84s/it]
                                                        

 34%|███▍      | 4980/14725 [2:33:56<4:59:16,  1.84s/it]
 34%|███▍      | 4981/14725 [2:33:58<4:57:04,  1.83s/it]
 34%|███▍      | 4982/14725 [2:34:00<4:58:54,  1.84s/it]
 34%|███▍      | 4983/14725 [2:34:02<4:56:32,  1.83s/it]
 34%|███▍      | 4984/14725 [2:34:04<4:55:04,  1.82s/it]
 34%
+0: {'loss': 0.8257, 'grad_norm': 0.7239640590309365, 'learning_rate': 2e-05, 'epoch': 0.34}
+0: |███▍      | 4985/14725 [2:34:05<4:57:15,  1.83s/it]
 34%|███▍      | 4986/14725 [2:34:07<4:56:36,  1.83s/it]
 34%|███▍      | 4987/14725 [2:34:09<4:52:27,  1.80s/it]
 34%|███▍      | 4988/14725 [2:34:11<4:51:30,  1.80s/it]
 34%|███▍      | 4989/14725 [2:34:13<4:49:17,  1.78s/it]
 34%|███▍      | 4990/14725 [2:34:14<4:51:51,  1.80s/it]
                                                        

 34%|███▍      | 4990/14725 [2:34:14<4:51:51,  1.80s/it]
 34%|███▍      | 4991/14725 [2:34:16<4:49:04,  1.78s/it]
 34%|███▍      | 4992/14725 [2:34:18<4:50:12,  1.79s/it]
 34%|███▍      | 4993/14725 [2:34:20<4:53:38,  1.81s/it]
 34%|███▍      | 4994/14725 [2:34:22<4:53:03,  1.81s/it]
 34%|███▍      | 4995/14725 [2:34:23<4:51:57,  1.80s/it]
 34%|███▍      | 4996/14725 [2:34:25<4:54:04,  1.81s/it]
 34%|███▍      | 4997/14725 [2:34:27<4:55:16,  1.82s/it]
 34%|███▍      | 4998/14725 [2:34:29<4:54:16,  1.82s
+0: {'loss': 0.8496, 'grad_norm': 0.8645911001654185, 'learning_rate': 2e-05, 'epoch': 0.34}
+0: {'loss': 0.833, 'grad_norm': 0.7433075038110905, 'learning_rate': 2e-05, 'epoch': 0.34}
+0: /it]
 34%|███▍      | 4999/14725 [2:34:31<4:55:20,  1.82s/it]
 34%|███▍      | 5000/14725 [2:34:32<4:52:52,  1.81s/it]
                                                        

 34%|███▍      | 5000/14725 [2:34:32<4:52:52,  1.81s/it]
 34%|███▍      | 5001/14725 [2:34:34<4:56:23,  1.83s/it]
 34%|███▍      | 5002/14725 [2:34:36<4:54:49,  1.82s/it]
 34%|███▍      | 5003/14725 [2:34:38<4:52:57,  1.81s/it]
 34%|███▍      | 5004/14725 [2:34:40<4:56:21,  1.83s/it]
 34%|███▍      | 5005/14725 [2:34:42<4:53:16,  1.81s/it]
 34%|███▍      | 5006/14725 [2:34:43<4:58:28,  1.84s/it]
 34%|███▍      | 5007/14725 [2:34:45<4:53:59,  1.82s/it]
 34%|███▍      | 5008/14725 [2:34:47<4:52:09,  1.80s/it]
 34%|███▍      | 5009/14725 [2:34:49<4:49:49,  1.79s/it]
 34%|███▍      | 5010/14725 [2:34:51<4:47:32,  1.78s/it]
                                                        

 34%|███▍      | 5010/14725 [2:34:51<4:47:32,  1.7
+0: {'loss': 0.8399, 'grad_norm': 0.8783053035203555, 'learning_rate': 2e-05, 'epoch': 0.34}
+0: 8s/it]
 34%|███▍      | 5011/14725 [2:34:52<4:48:59,  1.78s/it]
 34%|███▍      | 5012/14725 [2:34:54<4:47:58,  1.78s/it]
 34%|███▍      | 5013/14725 [2:34:56<4:47:01,  1.77s/it]
 34%|███▍      | 5014/14725 [2:34:58<4:46:16,  1.77s/it]
 34%|███▍      | 5015/14725 [2:34:59<4:49:12,  1.79s/it]
 34%|███▍      | 5016/14725 [2:35:01<4:50:04,  1.79s/it]
 34%|███▍      | 5017/14725 [2:35:03<4:49:57,  1.79s/it]
 34%|███▍      | 5018/14725 [2:35:05<4:50:56,  1.80s/it]
 34%|███▍      | 5019/14725 [2:35:07<4:50:34,  1.80s/it]
 34%|███▍      | 5020/14725 [2:35:08<4:50:09,  1.79s/it]
                                                        

 34%|███▍      | 5020/14725 [2:35:08<4:50:09,  1.79s/it]
 34%|███▍      | 5021/14725 [2:35:10<4:48:44,  1.79s/it]
 34%|███▍      | 5022/14725 [2:35:12<4:48:01,  1.78s/it]
 34%|███▍      | 5023/14725 [2:35:14<4:46:51,  1.77s/it]
 34%|███▍      | 5024/14725 [2:35:16<4:51
+0: {'loss': 0.8659, 'grad_norm': 0.8288706282360089, 'learning_rate': 2e-05, 'epoch': 0.34}
+0: :22,  1.80s/it]
 34%|███▍      | 5025/14725 [2:35:17<4:51:48,  1.80s/it]
 34%|███▍      | 5026/14725 [2:35:19<4:49:33,  1.79s/it]
 34%|███▍      | 5027/14725 [2:35:21<4:48:44,  1.79s/it]
 34%|███▍      | 5028/14725 [2:35:23<4:52:11,  1.81s/it]
 34%|███▍      | 5029/14725 [2:35:25<4:49:51,  1.79s/it]
 34%|███▍      | 5030/14725 [2:35:26<4:50:17,  1.80s/it]
                                                        

 34%|███▍      | 5030/14725 [2:35:26<4:50:17,  1.80s/it]
 34%|███▍      | 5031/14725 [2:35:28<4:57:55,  1.84s/it]
 34%|███▍      | 5032/14725 [2:35:30<4:59:18,  1.85s/it]
 34%|███▍      | 5033/14725 [2:35:32<4:55:08,  1.83s/it]
 34%|███▍      | 5034/14725 [2:35:34<4:52:59,  1.81s/it]
 34%|███▍      | 5035/14725 [2:35:36<4:54:03,  1.82s/it]
 34%|███▍      | 5036/14725 [2:35:37<4:54:53,  1.83s/it]
 34%|███▍      | 5037/14725 [2:35:39<4:57:21,  1.84s/it]
 34%|███▍      | 5038/14725 [2:3
+0: {'loss': 0.8565, 'grad_norm': 0.7902943060638457, 'learning_rate': 2e-05, 'epoch': 0.34}
+0: {'loss': 0.8372, 'grad_norm': 0.7171339383764342, 'learning_rate': 2e-05, 'epoch': 0.34}
+0: 5:41<4:53:34,  1.82s/it]
 34%|███▍      | 5039/14725 [2:35:43<4:52:41,  1.81s/it]
 34%|███▍      | 5040/14725 [2:35:45<4:55:38,  1.83s/it]
                                                        

 34%|███▍      | 5040/14725 [2:35:45<4:55:38,  1.83s/it]
 34%|███▍      | 5041/14725 [2:35:47<4:54:38,  1.83s/it]
 34%|███▍      | 5042/14725 [2:35:48<4:51:00,  1.80s/it]
 34%|███▍      | 5043/14725 [2:35:50<4:48:39,  1.79s/it]
 34%|███▍      | 5044/14725 [2:35:52<4:51:19,  1.81s/it]
 34%|███▍      | 5045/14725 [2:35:54<4:55:58,  1.83s/it]
 34%|███▍      | 5046/14725 [2:35:56<4:55:15,  1.83s/it]
 34%|███▍      | 5047/14725 [2:35:57<4:53:20,  1.82s/it]
 34%|███▍      | 5048/14725 [2:35:59<4:52:04,  1.81s/it]
 34%|███▍      | 5049/14725 [2:36:01<4:54:40,  1.83s/it]
 34%|███▍      | 5050/14725 [2:36:03<4:51:45,  1.81s/it]
                                                        

 34%|███▍      | 5050/14725 [2
+0: {'loss': 0.8516, 'grad_norm': 0.7458647167124022, 'learning_rate': 2e-05, 'epoch': 0.34}
+0: :36:03<4:51:45,  1.81s/it]
 34%|███▍      | 5051/14725 [2:36:05<4:51:14,  1.81s/it]
 34%|███▍      | 5052/14725 [2:36:06<4:49:40,  1.80s/it]
 34%|███▍      | 5053/14725 [2:36:08<4:50:42,  1.80s/it]
 34%|███▍      | 5054/14725 [2:36:10<4:53:48,  1.82s/it]
 34%|███▍      | 5055/14725 [2:36:12<4:50:39,  1.80s/it]
 34%|███▍      | 5056/14725 [2:36:14<4:48:01,  1.79s/it]
 34%|███▍      | 5057/14725 [2:36:15<4:47:56,  1.79s/it]
 34%|███▍      | 5058/14725 [2:36:17<4:53:34,  1.82s/it]
 34%|███▍      | 5059/14725 [2:36:19<4:52:41,  1.82s/it]
 34%|███▍      | 5060/14725 [2:36:21<4:57:11,  1.84s/it]
                                                        

 34%|███▍      | 5060/14725 [2:36:21<4:57:11,  1.84s/it]
 34%|███▍      | 5061/14725 [2:36:23<4:58:34,  1.85s/it]
 34%|███▍      | 5062/14725 [2:36:25<4:55:57,  1.84s/it]
 34%|███▍      | 5063/14725 [2:36:26<4:52:53,  1.82s/it]
 34%|███▍      | 5064
+0: {'loss': 0.8397, 'grad_norm': 0.7900175319571313, 'learning_rate': 2e-05, 'epoch': 0.34}
+0: /14725 [2:36:28<4:53:00,  1.82s/it]
 34%|███▍      | 5065/14725 [2:36:30<4:49:01,  1.80s/it]
 34%|███▍      | 5066/14725 [2:36:32<4:50:01,  1.80s/it]
 34%|███▍      | 5067/14725 [2:36:34<4:54:07,  1.83s/it]
 34%|███▍      | 5068/14725 [2:36:36<4:52:54,  1.82s/it]
 34%|███▍      | 5069/14725 [2:36:37<4:50:47,  1.81s/it]
 34%|███▍      | 5070/14725 [2:36:39<4:52:09,  1.82s/it]
                                                        

 34%|███▍      | 5070/14725 [2:36:39<4:52:09,  1.82s/it]
 34%|███▍      | 5071/14725 [2:36:41<4:50:16,  1.80s/it]
 34%|███▍      | 5072/14725 [2:36:43<4:51:52,  1.81s/it]
 34%|███▍      | 5073/14725 [2:36:45<4:50:08,  1.80s/it]
 34%|███▍      | 5074/14725 [2:36:46<4:51:31,  1.81s/it]
 34%|███▍      | 5075/14725 [2:36:48<4:50:01,  1.80s/it]
 34%|███▍      | 5076/14725 [2:36:50<4:50:06,  1.80s/it]
 34%|███▍      | 5077/14725 [2:36:52<4:49:20,  1.80s/it]
 34%|███▍   
+0: {'loss': 0.849, 'grad_norm': 0.733253813975833, 'learning_rate': 2e-05, 'epoch': 0.34}
+0: {'loss': 0.8344, 'grad_norm': 0.7736508387342322, 'learning_rate': 2e-05, 'epoch': 0.35}
+0:    | 5078/14725 [2:36:54<4:59:13,  1.86s/it]
 34%|███▍      | 5079/14725 [2:36:56<4:57:29,  1.85s/it]
 34%|███▍      | 5080/14725 [2:36:57<4:56:50,  1.85s/it]
                                                        

 34%|███▍      | 5080/14725 [2:36:57<4:56:50,  1.85s/it]
 35%|███▍      | 5081/14725 [2:36:59<4:54:01,  1.83s/it]
 35%|███▍      | 5082/14725 [2:37:01<4:56:46,  1.85s/it]
 35%|███▍      | 5083/14725 [2:37:03<4:53:32,  1.83s/it]
 35%|███▍      | 5084/14725 [2:37:05<4:53:12,  1.82s/it]
 35%|███▍      | 5085/14725 [2:37:06<4:50:10,  1.81s/it]
 35%|███▍      | 5086/14725 [2:37:08<4:53:55,  1.83s/it]
 35%|███▍      | 5087/14725 [2:37:10<4:51:32,  1.81s/it]
 35%|███▍      | 5088/14725 [2:37:12<4:54:59,  1.84s/it]
 35%|███▍      | 5089/14725 [2:37:14<4:50:37,  1.81s/it]
 35%|███▍      | 5090/14725 [2:37:15<4:48:35,  1.80s/it]
                                                        

 35%|███▍ 
+0: {'loss': 0.8456, 'grad_norm': 0.7418018136416791, 'learning_rate': 2e-05, 'epoch': 0.35}
+0:      | 5090/14725 [2:37:16<4:48:35,  1.80s/it]
 35%|███▍      | 5091/14725 [2:37:17<4:47:48,  1.79s/it]
 35%|███▍      | 5092/14725 [2:37:19<4:50:32,  1.81s/it]
 35%|███▍      | 5093/14725 [2:37:21<4:53:28,  1.83s/it]
 35%|███▍      | 5094/14725 [2:37:23<4:52:53,  1.82s/it]
 35%|███▍      | 5095/14725 [2:37:25<4:55:17,  1.84s/it]
 35%|███▍      | 5096/14725 [2:37:26<4:52:27,  1.82s/it]
 35%|███▍      | 5097/14725 [2:37:28<4:55:12,  1.84s/it]
 35%|███▍      | 5098/14725 [2:37:30<4:54:53,  1.84s/it]
 35%|███▍      | 5099/14725 [2:37:32<4:57:18,  1.85s/it]
 35%|███▍      | 5100/14725 [2:37:34<4:54:12,  1.83s/it]
                                                        

 35%|███▍      | 5100/14725 [2:37:34<4:54:12,  1.83s/it]
 35%|███▍      | 5101/14725 [2:37:36<4:51:12,  1.82s/it]
 35%|███▍      | 5102/14725 [2:37:37<4:51:17,  1.82s/it]
 35%|███▍      | 5103/14725 [2:37:39<4:51:50,  1.82s/it]
 35%|█�
+0: {'loss': 0.8366, 'grad_norm': 0.7375736439589405, 'learning_rate': 2e-05, 'epoch': 0.35}
+0: ��█▍      | 5104/14725 [2:37:41<4:50:29,  1.81s/it]
 35%|███▍      | 5105/14725 [2:37:43<4:51:00,  1.82s/it]
 35%|███▍      | 5106/14725 [2:37:45<4:47:38,  1.79s/it]
 35%|███▍      | 5107/14725 [2:37:46<4:46:19,  1.79s/it]
 35%|███▍      | 5108/14725 [2:37:48<4:44:00,  1.77s/it]
 35%|███▍      | 5109/14725 [2:37:50<4:45:01,  1.78s/it]
 35%|███▍      | 5110/14725 [2:37:52<4:46:33,  1.79s/it]
                                                        

 35%|███▍      | 5110/14725 [2:37:52<4:46:33,  1.79s/it]
 35%|███▍      | 5111/14725 [2:37:54<4:47:31,  1.79s/it]
 35%|███▍      | 5112/14725 [2:37:55<4:48:42,  1.80s/it]
 35%|███▍      | 5113/14725 [2:37:57<4:48:29,  1.80s/it]
 35%|███▍      | 5114/14725 [2:37:59<4:46:42,  1.79s/it]
 35%|███▍      | 5115/14725 [2:38:01<4:45:41,  1.78s/it]
 35%|███▍      | 5116/14725 [2:38:03<4:47:14,  1.79s/it]
 35%|███▍      | 5117/14725 [2:38:04<4:45:13,  1.78s/it]
+0: {'loss': 0.8368, 'grad_norm': 0.7597365700811732, 'learning_rate': 2e-05, 'epoch': 0.35}
+0: {'loss': 0.84, 'grad_norm': 0.8555817094127751, 'learning_rate': 2e-05, 'epoch': 0.35}
+0:  35%|███▍      | 5118/14725 [2:38:06<4:46:31,  1.79s/it]
 35%|███▍      | 5119/14725 [2:38:08<4:46:51,  1.79s/it]
 35%|███▍      | 5120/14725 [2:38:10<4:48:03,  1.80s/it]
                                                        

 35%|███▍      | 5120/14725 [2:38:10<4:48:03,  1.80s/it]
 35%|███▍      | 5121/14725 [2:38:12<4:49:35,  1.81s/it]
 35%|███▍      | 5122/14725 [2:38:13<4:50:13,  1.81s/it]
 35%|███▍      | 5123/14725 [2:38:15<4:46:18,  1.79s/it]
 35%|███▍      | 5124/14725 [2:38:17<4:47:37,  1.80s/it]
 35%|███▍      | 5125/14725 [2:38:19<4:46:41,  1.79s/it]
 35%|███▍      | 5126/14725 [2:38:21<5:03:54,  1.90s/it]
 35%|███▍      | 5127/14725 [2:38:23<4:59:09,  1.87s/it]
 35%|███▍      | 5128/14725 [2:38:24<4:56:21,  1.85s/it]
 35%|███▍      | 5129/14725 [2:38:26<4:56:11,  1.85s/it]
 35%|███▍      | 5130/14725 [2:38:28<4:59:03,  1.87s/it]
                                                        
+0: {'loss': 0.8361, 'grad_norm': 0.7073624076183623, 'learning_rate': 2e-05, 'epoch': 0.35}
+0: 

 35%|███▍      | 5130/14725 [2:38:28<4:59:03,  1.87s/it]
 35%|███▍      | 5131/14725 [2:38:30<4:55:02,  1.85s/it]
 35%|███▍      | 5132/14725 [2:38:32<4:53:24,  1.84s/it]
 35%|███▍      | 5133/14725 [2:38:34<4:55:43,  1.85s/it]
 35%|███▍      | 5134/14725 [2:38:36<4:54:46,  1.84s/it]
 35%|███▍      | 5135/14725 [2:38:37<4:52:34,  1.83s/it]
 35%|███▍      | 5136/14725 [2:38:39<4:55:52,  1.85s/it]
 35%|███▍      | 5137/14725 [2:38:41<4:55:54,  1.85s/it]
 35%|███▍      | 5138/14725 [2:38:43<4:51:15,  1.82s/it]
 35%|███▍      | 5139/14725 [2:38:45<4:55:44,  1.85s/it]
 35%|███▍      | 5140/14725 [2:38:47<4:54:09,  1.84s/it]
                                                        

 35%|███▍      | 5140/14725 [2:38:47<4:54:09,  1.84s/it]
 35%|███▍      | 5141/14725 [2:38:48<4:55:47,  1.85s/it]
 35%|███▍      | 5142/14725 [2:38:50<4:54:54,  1.85s/it]
 35%|███▍      | 5143/14725 [2:38:52<4:53:48, 
+0: {'loss': 0.8412, 'grad_norm': 0.7372231951923724, 'learning_rate': 2e-05, 'epoch': 0.35}
+0:  1.84s/it]
 35%|███▍      | 5144/14725 [2:38:54<4:51:19,  1.82s/it]
 35%|███▍      | 5145/14725 [2:38:56<4:48:55,  1.81s/it]
 35%|███▍      | 5146/14725 [2:38:58<4:52:50,  1.83s/it]
 35%|███▍      | 5147/14725 [2:38:59<4:51:37,  1.83s/it]
 35%|███▍      | 5148/14725 [2:39:01<4:49:53,  1.82s/it]
 35%|███▍      | 5149/14725 [2:39:03<4:47:46,  1.80s/it]
 35%|███▍      | 5150/14725 [2:39:05<4:46:38,  1.80s/it]
                                                        

 35%|███▍      | 5150/14725 [2:39:05<4:46:38,  1.80s/it]
 35%|███▍      | 5151/14725 [2:39:07<4:46:59,  1.80s/it]
 35%|███▍      | 5152/14725 [2:39:08<4:45:39,  1.79s/it]
 35%|███▍      | 5153/14725 [2:39:10<4:50:30,  1.82s/it]
 35%|███▌      | 5154/14725 [2:39:12<4:54:26,  1.85s/it]
 35%|███▌      | 5155/14725 [2:39:14<4:51:52,  1.83s/it]
 35%|███▌      | 5156/14725 [2:39:16<4:49:35,  1.82s/it]
 35%|███▌      | 5157/14725 [2:39:17<
+0: {'loss': 0.8409, 'grad_norm': 0.8053734738466908, 'learning_rate': 2e-05, 'epoch': 0.35}
+0: 4:46:47,  1.80s/it]
 35%|███▌      | 5158/14725 [2:39:19<4:45:20,  1.79s/it]
 35%|███▌      | 5159/14725 [2:39:21<4:48:48,  1.81s/it]
 35%|███▌      | 5160/14725 [2:39:23<4:47:24,  1.80s/it]
                                                        

 35%|███▌      | 5160/14725 [2:39:23<4:47:24,  1.80s/it]
 35%|███▌      | 5161/14725 [2:39:25<5:05:44,  1.92s/it]
 35%|███▌      | 5162/14725 [2:39:27<5:03:01,  1.90s/it]
 35%|███▌      | 5163/14725 [2:39:29<4:58:07,  1.87s/it]
 35%|███▌      | 5164/14725 [2:39:30<4:54:21,  1.85s/it]
 35%|███▌      | 5165/14725 [2:39:32<4:56:38,  1.86s/it]
 35%|███▌      | 5166/14725 [2:39:34<5:01:38,  1.89s/it]
 35%|███▌      | 5167/14725 [2:39:36<4:57:21,  1.87s/it]
 35%|███▌      | 5168/14725 [2:39:38<4:54:19,  1.85s/it]
 35%|███▌      | 5169/14725 [2:39:40<4:51:53,  1.83s/it]
 35%|███▌      | 5170/14725 [2:39:42<4:48:23,  1.81s/it]
                                    
+0: {'loss': 0.8395, 'grad_norm': 0.7789833349137197, 'learning_rate': 2e-05, 'epoch': 0.35}
+0: {'loss': 0.8352, 'grad_norm': 0.7200939057436266, 'learning_rate': 2e-05, 'epoch': 0.35}
+0:                     

 35%|███▌      | 5170/14725 [2:39:42<4:48:23,  1.81s/it]
 35%|███▌      | 5171/14725 [2:39:43<4:45:42,  1.79s/it]
 35%|███▌      | 5172/14725 [2:39:45<4:48:00,  1.81s/it]
 35%|███▌      | 5173/14725 [2:39:47<4:50:06,  1.82s/it]
 35%|███▌      | 5174/14725 [2:39:49<4:50:33,  1.83s/it]
 35%|███▌      | 5175/14725 [2:39:51<4:48:07,  1.81s/it]
 35%|███▌      | 5176/14725 [2:39:52<4:46:19,  1.80s/it]
 35%|███▌      | 5177/14725 [2:39:54<4:47:17,  1.81s/it]
 35%|███▌      | 5178/14725 [2:39:56<4:48:16,  1.81s/it]
 35%|███▌      | 5179/14725 [2:39:58<5:05:07,  1.92s/it]
 35%|███▌      | 5180/14725 [2:40:00<5:01:24,  1.89s/it]
                                                        

 35%|███▌      | 5180/14725 [2:40:00<5:01:24,  1.89s/it]
 35%|███▌      | 5181/14725 [2:40:02<4:56:08,  1.86s/it]
 35%|███▌      | 5182/14725 [2:40:04<5:05:20,  1.92s/it]
 35%|███▌      | 5183/1472
+0: {'loss': 0.8527, 'grad_norm': 0.778718139792205, 'learning_rate': 2e-05, 'epoch': 0.35}
+0: 5 [2:40:06<4:59:34,  1.88s/it]
 35%|███▌      | 5184/14725 [2:40:07<4:56:36,  1.87s/it]
 35%|███▌      | 5185/14725 [2:40:09<4:51:24,  1.83s/it]
 35%|███▌      | 5186/14725 [2:40:11<4:49:55,  1.82s/it]
 35%|███▌      | 5187/14725 [2:40:13<4:50:04,  1.82s/it]
 35%|███▌      | 5188/14725 [2:40:15<4:50:42,  1.83s/it]
 35%|███▌      | 5189/14725 [2:40:17<4:51:50,  1.84s/it]
 35%|███▌      | 5190/14725 [2:40:18<4:52:56,  1.84s/it]
                                                        

 35%|███▌      | 5190/14725 [2:40:18<4:52:56,  1.84s/it]
 35%|███▌      | 5191/14725 [2:40:20<4:51:20,  1.83s/it]
 35%|███▌      | 5192/14725 [2:40:22<4:48:09,  1.81s/it]
 35%|███▌      | 5193/14725 [2:40:24<4:47:25,  1.81s/it]
 35%|███▌      | 5194/14725 [2:40:26<4:45:34,  1.80s/it]
 35%|███▌      | 5195/14725 [2:40:27<4:48:09,  1.81s/it]
 35%|███▌      | 5196/14725 [2:40:30<5:07:51,  1.94s/it]
 35%|███▌      | 
+0: {'loss': 0.8236, 'grad_norm': 0.7327185056242778, 'learning_rate': 2e-05, 'epoch': 0.35}
+0: 5197/14725 [2:40:31<4:58:29,  1.88s/it]
 35%|███▌      | 5198/14725 [2:40:33<4:54:25,  1.85s/it]
 35%|███▌      | 5199/14725 [2:40:35<4:53:45,  1.85s/it]
 35%|███▌      | 5200/14725 [2:40:37<4:49:40,  1.82s/it]
                                                        

 35%|███▌      | 5200/14725 [2:40:37<4:49:40,  1.82s/it]
 35%|███▌      | 5201/14725 [2:40:39<4:50:22,  1.83s/it]
 35%|███▌      | 5202/14725 [2:40:40<4:47:42,  1.81s/it]
 35%|███▌      | 5203/14725 [2:40:42<4:50:36,  1.83s/it]
 35%|███▌      | 5204/14725 [2:40:44<4:48:10,  1.82s/it]
 35%|███▌      | 5205/14725 [2:40:46<4:49:41,  1.83s/it]
 35%|███▌      | 5206/14725 [2:40:48<4:50:58,  1.83s/it]
 35%|███▌      | 5207/14725 [2:40:50<4:52:13,  1.84s/it]
 35%|███▌      | 5208/14725 [2:40:51<4:49:57,  1.83s/it]
 35%|███▌      | 5209/14725 [2:40:53<4:47:40,  1.81s/it]
 35%|███▌      | 5210/14725 [2:40:55<4:46:11,  1.80s/it]
                
+0: {'loss': 0.8346, 'grad_norm': 0.735591603726033, 'learning_rate': 2e-05, 'epoch': 0.35}
+0: {'loss': 0.8488, 'grad_norm': 0.6982830633963028, 'learning_rate': 2e-05, 'epoch': 0.35}
+0:                                         

 35%|███▌      | 5210/14725 [2:40:55<4:46:11,  1.80s/it]
 35%|███▌      | 5211/14725 [2:40:57<4:48:50,  1.82s/it]
 35%|███▌      | 5212/14725 [2:40:59<4:46:35,  1.81s/it]
 35%|███▌      | 5213/14725 [2:41:00<4:44:46,  1.80s/it]
 35%|███▌      | 5214/14725 [2:41:02<4:48:17,  1.82s/it]
 35%|███▌      | 5215/14725 [2:41:04<4:46:26,  1.81s/it]
 35%|███▌      | 5216/14725 [2:41:06<4:50:35,  1.83s/it]
 35%|███▌      | 5217/14725 [2:41:08<4:47:23,  1.81s/it]
 35%|███▌      | 5218/14725 [2:41:10<4:48:35,  1.82s/it]
 35%|███▌      | 5219/14725 [2:41:11<4:52:21,  1.85s/it]
 35%|███▌      | 5220/14725 [2:41:13<4:50:28,  1.83s/it]
                                                        

 35%|███▌      | 5220/14725 [2:41:13<4:50:28,  1.83s/it]
 35%|███▌      | 5221/14725 [2:41:15<4:49:04,  1.82s/it]
 35%|███▌      | 5222/14725 [2:41:17<4:49:48,  1.83s/it]
 35%|███
+0: {'loss': 0.8381, 'grad_norm': 0.7138915543695754, 'learning_rate': 2e-05, 'epoch': 0.36}
+0: ▌      | 5223/14725 [2:41:19<4:46:13,  1.81s/it]
 35%|███▌      | 5224/14725 [2:41:20<4:46:38,  1.81s/it]
 35%|███▌      | 5225/14725 [2:41:22<4:49:15,  1.83s/it]
 35%|███▌      | 5226/14725 [2:41:24<4:48:50,  1.82s/it]
 35%|███▌      | 5227/14725 [2:41:26<4:45:55,  1.81s/it]
 36%|███▌      | 5228/14725 [2:41:28<4:45:49,  1.81s/it]
 36%|███▌      | 5229/14725 [2:41:30<4:47:31,  1.82s/it]
 36%|███▌      | 5230/14725 [2:41:31<4:47:55,  1.82s/it]
                                                        

 36%|███▌      | 5230/14725 [2:41:31<4:47:55,  1.82s/it]
 36%|███▌      | 5231/14725 [2:41:33<4:49:11,  1.83s/it]
 36%|███▌      | 5232/14725 [2:41:35<4:50:40,  1.84s/it]
 36%|███▌      | 5233/14725 [2:41:37<4:53:09,  1.85s/it]
 36%|███▌      | 5234/14725 [2:41:39<4:51:11,  1.84s/it]
 36%|███▌      | 5235/14725 [2:41:41<4:46:31,  1.81s/it]
 36%|███▌      | 5236/14725 [2:41:42<4:47:56,  1.82s/it]
 36%|
+0: {'loss': 0.8529, 'grad_norm': 0.7117123485242156, 'learning_rate': 2e-05, 'epoch': 0.36}
+0: ███▌      | 5237/14725 [2:41:44<4:50:10,  1.84s/it]
 36%|███▌      | 5238/14725 [2:41:46<4:50:00,  1.83s/it]
 36%|███▌      | 5239/14725 [2:41:48<4:49:22,  1.83s/it]
 36%|███▌      | 5240/14725 [2:41:50<4:47:58,  1.82s/it]
                                                        

 36%|███▌      | 5240/14725 [2:41:50<4:47:58,  1.82s/it]
 36%|███▌      | 5241/14725 [2:41:51<4:46:20,  1.81s/it]
 36%|███▌      | 5242/14725 [2:41:53<4:43:47,  1.80s/it]
 36%|███▌      | 5243/14725 [2:41:55<4:44:48,  1.80s/it]
 36%|███▌      | 5244/14725 [2:41:57<4:43:58,  1.80s/it]
 36%|███▌      | 5245/14725 [2:41:59<4:43:30,  1.79s/it]
 36%|███▌      | 5246/14725 [2:42:00<4:44:42,  1.80s/it]
 36%|███▌      | 5247/14725 [2:42:02<4:46:05,  1.81s/it]
 36%|███▌      | 5248/14725 [2:42:04<4:43:38,  1.80s/it]
 36%|███▌      | 5249/14725 [2:42:06<4:47:20,  1.82s/it]
 36%|███▌      | 5250/14725 [2:42:08<4:45:31,  1.81s/
+0: {'loss': 0.8303, 'grad_norm': 0.791531164221219, 'learning_rate': 2e-05, 'epoch': 0.36}
+0: {'loss': 0.8501, 'grad_norm': 0.8871301482302663, 'learning_rate': 2e-05, 'epoch': 0.36}
+0: it]
                                                        

 36%|███▌      | 5250/14725 [2:42:08<4:45:31,  1.81s/it]
 36%|███▌      | 5251/14725 [2:42:10<4:53:53,  1.86s/it]
 36%|███▌      | 5252/14725 [2:42:11<4:49:42,  1.83s/it]
 36%|███▌      | 5253/14725 [2:42:13<4:46:52,  1.82s/it]
 36%|███▌      | 5254/14725 [2:42:15<4:44:53,  1.80s/it]
 36%|███▌      | 5255/14725 [2:42:17<4:47:50,  1.82s/it]
 36%|███▌      | 5256/14725 [2:42:19<4:50:37,  1.84s/it]
 36%|███▌      | 5257/14725 [2:42:21<4:51:20,  1.85s/it]
 36%|███▌      | 5258/14725 [2:42:22<4:47:39,  1.82s/it]
 36%|███▌      | 5259/14725 [2:42:24<4:46:43,  1.82s/it]
 36%|███▌      | 5260/14725 [2:42:26<4:45:47,  1.81s/it]
                                                        

 36%|███▌      | 5260/14725 [2:42:26<4:45:47,  1.81s/it]
 36%|███▌      | 5261/14725 [2:42:28<4:44:42,  1.80s/it]
 36%|███▌      | 5262/14725 [2:42:30<4:48:09,  1.83
+0: {'loss': 0.8258, 'grad_norm': 0.7197122913559522, 'learning_rate': 2e-05, 'epoch': 0.36}
+0: s/it]
 36%|███▌      | 5263/14725 [2:42:31<4:45:43,  1.81s/it]
 36%|███▌      | 5264/14725 [2:42:33<4:43:16,  1.80s/it]
 36%|███▌      | 5265/14725 [2:42:35<4:43:44,  1.80s/it]
 36%|███▌      | 5266/14725 [2:42:37<4:43:43,  1.80s/it]
 36%|███▌      | 5267/14725 [2:42:39<4:42:57,  1.80s/it]
 36%|███▌      | 5268/14725 [2:42:40<4:40:45,  1.78s/it]
 36%|███▌      | 5269/14725 [2:42:42<4:42:40,  1.79s/it]
 36%|███▌      | 5270/14725 [2:42:44<4:44:35,  1.81s/it]
                                                        

 36%|███▌      | 5270/14725 [2:42:44<4:44:35,  1.81s/it]
 36%|███▌      | 5271/14725 [2:42:46<4:45:49,  1.81s/it]
 36%|███▌      | 5272/14725 [2:42:48<4:47:34,  1.83s/it]
 36%|███▌      | 5273/14725 [2:42:49<4:47:09,  1.82s/it]
 36%|███▌      | 5274/14725 [2:42:51<4:43:17,  1.80s/it]
 36%|███▌      | 5275/14725 [2:42:53<4:46:21,  1.82s/it]
 36%|███▌      | 5276/14725 [2:42:55<4:44:
+0: {'loss': 0.8405, 'grad_norm': 0.7497051678424714, 'learning_rate': 2e-05, 'epoch': 0.36}
+0: 42,  1.81s/it]
 36%|███▌      | 5277/14725 [2:42:57<4:41:50,  1.79s/it]
 36%|███▌      | 5278/14725 [2:42:58<4:41:47,  1.79s/it]
 36%|███▌      | 5279/14725 [2:43:00<4:43:16,  1.80s/it]
 36%|███▌      | 5280/14725 [2:43:02<4:41:46,  1.79s/it]
                                                        

 36%|███▌      | 5280/14725 [2:43:02<4:41:46,  1.79s/it]
 36%|███▌      | 5281/14725 [2:43:04<4:41:36,  1.79s/it]
 36%|███▌      | 5282/14725 [2:43:06<4:40:26,  1.78s/it]
 36%|███▌      | 5283/14725 [2:43:07<4:39:08,  1.77s/it]
 36%|███▌      | 5284/14725 [2:43:09<4:39:34,  1.78s/it]
 36%|███▌      | 5285/14725 [2:43:11<4:40:32,  1.78s/it]
 36%|███▌      | 5286/14725 [2:43:13<4:40:09,  1.78s/it]
 36%|███▌      | 5287/14725 [2:43:15<5:11:50,  1.98s/it]
 36%|███▌      | 5288/14725 [2:43:17<5:10:20,  1.97s/it]
 36%|███▌      | 5289/14725 [2:43:19<5:02:57,  1.93s/it]
 36%|███▌      | 5290/14725 [2:43
+0: {'loss': 0.8342, 'grad_norm': 0.7380336333066729, 'learning_rate': 2e-05, 'epoch': 0.36}
+0: {'loss': 0.8126, 'grad_norm': 0.7174770627031654, 'learning_rate': 2e-05, 'epoch': 0.36}
+0: :21<4:58:00,  1.90s/it]
                                                        

 36%|███▌      | 5290/14725 [2:43:21<4:58:00,  1.90s/it]
 36%|███▌      | 5291/14725 [2:43:23<4:54:03,  1.87s/it]
 36%|███▌      | 5292/14725 [2:43:24<4:51:28,  1.85s/it]
 36%|███▌      | 5293/14725 [2:43:26<4:51:22,  1.85s/it]
 36%|███▌      | 5294/14725 [2:43:28<4:48:43,  1.84s/it]
 36%|███▌      | 5295/14725 [2:43:30<4:46:23,  1.82s/it]
 36%|███▌      | 5296/14725 [2:43:32<4:46:09,  1.82s/it]
 36%|███▌      | 5297/14725 [2:43:33<4:50:19,  1.85s/it]
 36%|███▌      | 5298/14725 [2:43:35<4:50:21,  1.85s/it]
 36%|███▌      | 5299/14725 [2:43:37<4:46:14,  1.82s/it]
 36%|███▌      | 5300/14725 [2:43:39<4:44:32,  1.81s/it]
                                                        

 36%|███▌      | 5300/14725 [2:43:39<4:44:32,  1.81s/it]
 36%|███▌      | 5301/14725 [2:43:41<4:45:12,  1.82s/it]
 36%|███▌      | 5302/14725 [2:
+0: {'loss': 0.8257, 'grad_norm': 0.7268195865302866, 'learning_rate': 2e-05, 'epoch': 0.36}
+0: 43:43<4:44:25,  1.81s/it]
 36%|███▌      | 5303/14725 [2:43:44<4:42:56,  1.80s/it]
 36%|███▌      | 5304/14725 [2:43:46<4:43:59,  1.81s/it]
 36%|███▌      | 5305/14725 [2:43:48<4:42:48,  1.80s/it]
 36%|███▌      | 5306/14725 [2:43:50<4:41:51,  1.80s/it]
 36%|███▌      | 5307/14725 [2:43:52<4:42:57,  1.80s/it]
 36%|███▌      | 5308/14725 [2:43:53<4:43:13,  1.80s/it]
 36%|███▌      | 5309/14725 [2:43:55<4:41:56,  1.80s/it]
 36%|███▌      | 5310/14725 [2:43:57<4:41:28,  1.79s/it]
                                                        

 36%|███▌      | 5310/14725 [2:43:57<4:41:28,  1.79s/it]
 36%|███▌      | 5311/14725 [2:43:59<4:44:00,  1.81s/it]
 36%|███▌      | 5312/14725 [2:44:00<4:41:26,  1.79s/it]
 36%|███▌      | 5313/14725 [2:44:02<4:42:33,  1.80s/it]
 36%|███▌      | 5314/14725 [2:44:04<4:42:25,  1.80s/it]
 36%|███▌      | 5315/14725 [2:44:06<4:43:34,  1.81s/it]
 36%|███▌      | 5316/
+0: {'loss': 0.854, 'grad_norm': 0.7430817270476819, 'learning_rate': 2e-05, 'epoch': 0.36}
+0: 14725 [2:44:08<4:44:51,  1.82s/it]
 36%|███▌      | 5317/14725 [2:44:10<4:44:44,  1.82s/it]
 36%|███▌      | 5318/14725 [2:44:11<4:46:48,  1.83s/it]
 36%|███▌      | 5319/14725 [2:44:13<4:43:10,  1.81s/it]
 36%|███▌      | 5320/14725 [2:44:15<4:43:53,  1.81s/it]
                                                        

 36%|███▌      | 5320/14725 [2:44:15<4:43:53,  1.81s/it]
 36%|███▌      | 5321/14725 [2:44:17<4:42:51,  1.80s/it]
 36%|███▌      | 5322/14725 [2:44:19<4:42:56,  1.81s/it]
 36%|███▌      | 5323/14725 [2:44:20<4:42:52,  1.81s/it]
 36%|███▌      | 5324/14725 [2:44:22<4:40:48,  1.79s/it]
 36%|███▌      | 5325/14725 [2:44:24<4:42:59,  1.81s/it]
 36%|███▌      | 5326/14725 [2:44:26<4:43:36,  1.81s/it]
 36%|███▌      | 5327/14725 [2:44:28<4:42:34,  1.80s/it]
 36%|███▌      | 5328/14725 [2:44:29<4:41:28,  1.80s/it]
 36%|███▌      | 5329/14725 [2:44:31<4:42:47,  1.81s/it]
 36%|███▌    
+0: {'loss': 0.8295, 'grad_norm': 0.7319291895806966, 'learning_rate': 2e-05, 'epoch': 0.36}
+0: {'loss': 0.8311, 'grad_norm': 0.74838728963627, 'learning_rate': 2e-05, 'epoch': 0.36}
+0:   | 5330/14725 [2:44:33<4:43:41,  1.81s/it]
                                                        

 36%|███▌      | 5330/14725 [2:44:33<4:43:41,  1.81s/it]
 36%|███▌      | 5331/14725 [2:44:35<4:43:10,  1.81s/it]
 36%|███▌      | 5332/14725 [2:44:37<4:42:04,  1.80s/it]
 36%|███▌      | 5333/14725 [2:44:38<4:42:56,  1.81s/it]
 36%|███▌      | 5334/14725 [2:44:40<4:43:15,  1.81s/it]
 36%|███▌      | 5335/14725 [2:44:42<4:41:31,  1.80s/it]
 36%|███▌      | 5336/14725 [2:44:44<4:42:01,  1.80s/it]
 36%|███▌      | 5337/14725 [2:44:46<4:41:52,  1.80s/it]
 36%|███▋      | 5338/14725 [2:44:47<4:41:19,  1.80s/it]
 36%|███▋      | 5339/14725 [2:44:49<4:42:53,  1.81s/it]
 36%|███▋      | 5340/14725 [2:44:51<4:40:30,  1.79s/it]
                                                        

 36%|███▋      | 5340/14725 [2:44:51<4:40:30,  1.79s/it]
 36%|███▋      | 5341/14725 [2:44:53<4:44:02,  1.82s/it]
 36%|███▋  
+0: {'loss': 0.8262, 'grad_norm': 0.8525578646123835, 'learning_rate': 2e-05, 'epoch': 0.36}
+0:     | 5342/14725 [2:44:55<4:44:33,  1.82s/it]
 36%|███▋      | 5343/14725 [2:44:57<4:41:50,  1.80s/it]
 36%|███▋      | 5344/14725 [2:44:58<4:40:52,  1.80s/it]
 36%|███▋      | 5345/14725 [2:45:00<4:42:41,  1.81s/it]
 36%|███▋      | 5346/14725 [2:45:02<4:48:13,  1.84s/it]
 36%|███▋      | 5347/14725 [2:45:04<4:47:06,  1.84s/it]
 36%|███▋      | 5348/14725 [2:45:06<4:47:30,  1.84s/it]
 36%|███▋      | 5349/14725 [2:45:08<4:47:07,  1.84s/it]
 36%|███▋      | 5350/14725 [2:45:09<4:47:14,  1.84s/it]
                                                        

 36%|███▋      | 5350/14725 [2:45:09<4:47:14,  1.84s/it]
 36%|███▋      | 5351/14725 [2:45:11<4:52:45,  1.87s/it]
 36%|███▋      | 5352/14725 [2:45:13<4:52:53,  1.87s/it]
 36%|███▋      | 5353/14725 [2:45:15<4:49:21,  1.85s/it]
 36%|███▋      | 5354/14725 [2:45:17<4:46:21,  1.83s/it]
 36%|███▋      | 5355/14725 [2:45:19<4:51:39,  1.87s/it]
 36%|█�
+0: {'loss': 0.8626, 'grad_norm': 0.6902223195025867, 'learning_rate': 2e-05, 'epoch': 0.36}
+0: �█▋      | 5356/14725 [2:45:21<4:49:35,  1.85s/it]
 36%|███▋      | 5357/14725 [2:45:22<4:51:39,  1.87s/it]
 36%|███▋      | 5358/14725 [2:45:24<4:48:50,  1.85s/it]
 36%|███▋      | 5359/14725 [2:45:26<4:46:39,  1.84s/it]
 36%|███▋      | 5360/14725 [2:45:28<4:47:22,  1.84s/it]
                                                        

 36%|███▋      | 5360/14725 [2:45:28<4:47:22,  1.84s/it]
 36%|███▋      | 5361/14725 [2:45:30<4:45:34,  1.83s/it]
 36%|███▋      | 5362/14725 [2:45:32<4:43:37,  1.82s/it]
 36%|███▋      | 5363/14725 [2:45:33<4:43:11,  1.81s/it]
 36%|███▋      | 5364/14725 [2:45:35<4:42:36,  1.81s/it]
 36%|███▋      | 5365/14725 [2:45:37<4:45:25,  1.83s/it]
 36%|███▋      | 5366/14725 [2:45:39<4:43:46,  1.82s/it]
 36%|███▋      | 5367/14725 [2:45:41<4:44:53,  1.83s/it]
 36%|███▋      | 5368/14725 [2:45:42<4:43:48,  1.82s/it]
 36%|███▋      | 5369/14725 [2:45:44<4:44:18,  1.82s/it]
 
+0: {'loss': 0.8295, 'grad_norm': 0.7276729027041637, 'learning_rate': 2e-05, 'epoch': 0.36}
+0: {'loss': 0.8332, 'grad_norm': 0.7622669938712165, 'learning_rate': 2e-05, 'epoch': 0.37}
+0: 36%|███▋      | 5370/14725 [2:45:46<4:43:21,  1.82s/it]
                                                        

 36%|███▋      | 5370/14725 [2:45:46<4:43:21,  1.82s/it]
 36%|███▋      | 5371/14725 [2:45:48<4:43:42,  1.82s/it]
 36%|███▋      | 5372/14725 [2:45:50<4:43:30,  1.82s/it]
 36%|███▋      | 5373/14725 [2:45:52<4:40:27,  1.80s/it]
 36%|███▋      | 5374/14725 [2:45:53<4:38:37,  1.79s/it]
 37%|███▋      | 5375/14725 [2:45:55<4:37:16,  1.78s/it]
 37%|███▋      | 5376/14725 [2:45:57<4:37:02,  1.78s/it]
 37%|███▋      | 5377/14725 [2:45:59<4:38:20,  1.79s/it]
 37%|███▋      | 5378/14725 [2:46:00<4:40:57,  1.80s/it]
 37%|███▋      | 5379/14725 [2:46:02<4:40:40,  1.80s/it]
 37%|███▋      | 5380/14725 [2:46:04<4:41:12,  1.81s/it]
                                                        

 37%|███▋      | 5380/14725 [2:46:04<4:41:12,  1.81s/it]
 37%|███▋      | 5381/14725 [2:46:06<4:40:04,  1.80s/it]
+0: {'loss': 0.8352, 'grad_norm': 0.7898824028090227, 'learning_rate': 2e-05, 'epoch': 0.37}
+0: 
 37%|███▋      | 5382/14725 [2:46:08<4:38:45,  1.79s/it]
 37%|███▋      | 5383/14725 [2:46:09<4:43:00,  1.82s/it]
 37%|███▋      | 5384/14725 [2:46:11<4:44:50,  1.83s/it]
 37%|███▋      | 5385/14725 [2:46:13<4:44:19,  1.83s/it]
 37%|███▋      | 5386/14725 [2:46:15<4:41:48,  1.81s/it]
 37%|███▋      | 5387/14725 [2:46:17<4:38:57,  1.79s/it]
 37%|███▋      | 5388/14725 [2:46:19<4:42:59,  1.82s/it]
 37%|███▋      | 5389/14725 [2:46:20<4:40:10,  1.80s/it]
 37%|███▋      | 5390/14725 [2:46:22<4:40:27,  1.80s/it]
                                                        

 37%|███▋      | 5390/14725 [2:46:22<4:40:27,  1.80s/it]
 37%|███▋      | 5391/14725 [2:46:24<4:38:41,  1.79s/it]
 37%|███▋      | 5392/14725 [2:46:26<4:39:31,  1.80s/it]
 37%|███▋      | 5393/14725 [2:46:27<4:37:35,  1.78s/it]
 37%|███▋      | 5394/14725 [2:46:29<4:40:23,  1.80s/it]
 37%|███▋      | 5395/14725 [2:46:31<4:39:03,  
+0: {'loss': 0.8303, 'grad_norm': 0.7132341223254809, 'learning_rate': 2e-05, 'epoch': 0.37}
+0: 1.79s/it]
 37%|███▋      | 5396/14725 [2:46:33<4:41:04,  1.81s/it]
 37%|███▋      | 5397/14725 [2:46:35<4:41:19,  1.81s/it]
 37%|███▋      | 5398/14725 [2:46:37<4:41:15,  1.81s/it]
 37%|███▋      | 5399/14725 [2:46:38<4:38:43,  1.79s/it]
 37%|███▋      | 5400/14725 [2:46:40<4:43:51,  1.83s/it]
                                                        

 37%|███▋      | 5400/14725 [2:46:40<4:43:51,  1.83s/it]
 37%|███▋      | 5401/14725 [2:46:42<4:41:08,  1.81s/it]
 37%|███▋      | 5402/14725 [2:46:44<4:43:55,  1.83s/it]
 37%|███▋      | 5403/14725 [2:46:46<4:42:38,  1.82s/it]
 37%|███▋      | 5404/14725 [2:46:47<4:41:48,  1.81s/it]
 37%|███▋      | 5405/14725 [2:46:49<4:38:57,  1.80s/it]
 37%|███▋      | 5406/14725 [2:46:51<4:40:28,  1.81s/it]
 37%|███▋      | 5407/14725 [2:46:53<4:40:46,  1.81s/it]
 37%|███▋      | 5408/14725 [2:46:55<4:39:21,  1.80s/it]
 37%|███▋      | 5409/14725 [2:46:56<4
+0: {'loss': 0.8332, 'grad_norm': 0.7339831636980334, 'learning_rate': 2e-05, 'epoch': 0.37}
+0: {'loss': 0.8153, 'grad_norm': 0.7351192454204599, 'learning_rate': 2e-05, 'epoch': 0.37}
+0: :42:03,  1.82s/it]
 37%|███▋      | 5410/14725 [2:46:58<4:43:16,  1.82s/it]
                                                        

 37%|███▋      | 5410/14725 [2:46:58<4:43:16,  1.82s/it]
 37%|███▋      | 5411/14725 [2:47:00<4:46:49,  1.85s/it]
 37%|███▋      | 5412/14725 [2:47:02<4:45:53,  1.84s/it]
 37%|███▋      | 5413/14725 [2:47:04<4:42:28,  1.82s/it]
 37%|███▋      | 5414/14725 [2:47:06<4:39:01,  1.80s/it]
 37%|███▋      | 5415/14725 [2:47:07<4:39:42,  1.80s/it]
 37%|███▋      | 5416/14725 [2:47:09<4:38:12,  1.79s/it]
 37%|███▋      | 5417/14725 [2:47:11<4:37:08,  1.79s/it]
 37%|███▋      | 5418/14725 [2:47:13<4:36:08,  1.78s/it]
 37%|███▋      | 5419/14725 [2:47:15<4:37:47,  1.79s/it]
 37%|███▋      | 5420/14725 [2:47:16<4:37:12,  1.79s/it]
                                                        

 37%|███▋      | 5420/14725 [2:47:16<4:37:12,  1.79s/it]
 37%|███▋      | 5421/14725 [2:47:18
+0: {'loss': 0.8415, 'grad_norm': 0.7389979875636221, 'learning_rate': 2e-05, 'epoch': 0.37}
+0: <4:39:09,  1.80s/it]
 37%|███▋      | 5422/14725 [2:47:20<4:42:00,  1.82s/it]
 37%|███▋      | 5423/14725 [2:47:22<4:41:40,  1.82s/it]
 37%|███▋      | 5424/14725 [2:47:24<4:40:04,  1.81s/it]
 37%|███▋      | 5425/14725 [2:47:25<4:38:33,  1.80s/it]
 37%|███▋      | 5426/14725 [2:47:27<4:37:52,  1.79s/it]
 37%|███▋      | 5427/14725 [2:47:29<4:41:04,  1.81s/it]
 37%|███▋      | 5428/14725 [2:47:31<4:41:17,  1.82s/it]
 37%|███▋      | 5429/14725 [2:47:33<4:40:20,  1.81s/it]
 37%|███▋      | 5430/14725 [2:47:34<4:40:43,  1.81s/it]
                                                        

 37%|███▋      | 5430/14725 [2:47:34<4:40:43,  1.81s/it]
 37%|███▋      | 5431/14725 [2:47:36<4:39:42,  1.81s/it]
 37%|███▋      | 5432/14725 [2:47:38<4:38:26,  1.80s/it]
 37%|███▋      | 5433/14725 [2:47:40<4:35:48,  1.78s/it]
 37%|███▋      | 5434/14725 [2:47:42<4:35:04,  1.78s/it]
 37%|███▋      | 5435/14725
+0: {'loss': 0.8275, 'grad_norm': 0.749223074345115, 'learning_rate': 2e-05, 'epoch': 0.37}
+0:  [2:47:43<4:35:53,  1.78s/it]
 37%|███▋      | 5436/14725 [2:47:45<4:33:30,  1.77s/it]
 37%|███▋      | 5437/14725 [2:47:47<4:35:24,  1.78s/it]
 37%|███▋      | 5438/14725 [2:47:49<4:34:28,  1.77s/it]
 37%|███▋      | 5439/14725 [2:47:50<4:38:35,  1.80s/it]
 37%|███▋      | 5440/14725 [2:47:52<4:36:48,  1.79s/it]
                                                        

 37%|███▋      | 5440/14725 [2:47:52<4:36:48,  1.79s/it]
 37%|███▋      | 5441/14725 [2:47:54<4:39:20,  1.81s/it]
 37%|███▋      | 5442/14725 [2:47:56<4:38:52,  1.80s/it]
 37%|███▋      | 5443/14725 [2:47:58<4:41:40,  1.82s/it]
 37%|███▋      | 5444/14725 [2:48:00<4:40:12,  1.81s/it]
 37%|███▋      | 5445/14725 [2:48:01<4:39:16,  1.81s/it]
 37%|███▋      | 5446/14725 [2:48:03<4:38:03,  1.80s/it]
 37%|███▋      | 5447/14725 [2:48:05<4:35:11,  1.78s/it]
 37%|███▋      | 5448/14725 [2:48:07<4:36:11,  1.79s/it]
 37%|███▋      | 5
+0: {'loss': 0.8297, 'grad_norm': 0.7261993276449172, 'learning_rate': 2e-05, 'epoch': 0.37}
+0: {'loss': 0.8402, 'grad_norm': 0.7108190794483472, 'learning_rate': 2e-05, 'epoch': 0.37}
+0: 449/14725 [2:48:09<4:47:43,  1.86s/it]
 37%|███▋      | 5450/14725 [2:48:10<4:44:26,  1.84s/it]
                                                        

 37%|███▋      | 5450/14725 [2:48:10<4:44:26,  1.84s/it]
 37%|███▋      | 5451/14725 [2:48:12<4:45:39,  1.85s/it]
 37%|███▋      | 5452/14725 [2:48:14<4:41:45,  1.82s/it]
 37%|███▋      | 5453/14725 [2:48:16<4:41:25,  1.82s/it]
 37%|███▋      | 5454/14725 [2:48:18<4:39:58,  1.81s/it]
 37%|███▋      | 5455/14725 [2:48:20<4:39:36,  1.81s/it]
 37%|███▋      | 5456/14725 [2:48:21<4:42:14,  1.83s/it]
 37%|███▋      | 5457/14725 [2:48:23<4:45:26,  1.85s/it]
 37%|███▋      | 5458/14725 [2:48:25<4:43:28,  1.84s/it]
 37%|███▋      | 5459/14725 [2:48:27<4:42:55,  1.83s/it]
 37%|███▋      | 5460/14725 [2:48:29<4:40:36,  1.82s/it]
                                                        

 37%|███▋      | 5460/14725 [2:48:29<4:40:36,  1.82s/it]
 37%|███▋      |
+0: {'loss': 0.8315, 'grad_norm': 0.7428608363605325, 'learning_rate': 2e-05, 'epoch': 0.37}
+0:  5461/14725 [2:48:30<4:37:18,  1.80s/it]
 37%|███▋      | 5462/14725 [2:48:32<4:37:39,  1.80s/it]
 37%|███▋      | 5463/14725 [2:48:34<4:37:24,  1.80s/it]
 37%|███▋      | 5464/14725 [2:48:36<4:39:05,  1.81s/it]
 37%|███▋      | 5465/14725 [2:48:38<4:40:01,  1.81s/it]
 37%|███▋      | 5466/14725 [2:48:39<4:36:30,  1.79s/it]
 37%|███▋      | 5467/14725 [2:48:41<4:34:48,  1.78s/it]
 37%|███▋      | 5468/14725 [2:48:43<4:34:14,  1.78s/it]
 37%|███▋      | 5469/14725 [2:48:45<4:34:30,  1.78s/it]
 37%|███▋      | 5470/14725 [2:48:46<4:32:53,  1.77s/it]
                                                        

 37%|███▋      | 5470/14725 [2:48:46<4:32:53,  1.77s/it]
 37%|███▋      | 5471/14725 [2:48:48<4:36:04,  1.79s/it]
 37%|███▋      | 5472/14725 [2:48:50<4:35:16,  1.78s/it]
 37%|███▋      | 5473/14725 [2:48:52<4:36:22,  1.79s/it]
 37%|███▋      | 5474/14725 [2:48:54<4:34:24,  1.78s/it]
 37%|███�
+0: {'loss': 0.8304, 'grad_norm': 0.8844438671079685, 'learning_rate': 2e-05, 'epoch': 0.37}
+0: ��      | 5475/14725 [2:48:56<4:39:35,  1.81s/it]
 37%|███▋      | 5476/14725 [2:48:57<4:39:37,  1.81s/it]
 37%|███▋      | 5477/14725 [2:48:59<4:41:37,  1.83s/it]
 37%|███▋      | 5478/14725 [2:49:01<4:41:10,  1.82s/it]
 37%|███▋      | 5479/14725 [2:49:03<4:39:07,  1.81s/it]
 37%|███▋      | 5480/14725 [2:49:05<4:43:27,  1.84s/it]
                                                        

 37%|███▋      | 5480/14725 [2:49:05<4:43:27,  1.84s/it]
 37%|███▋      | 5481/14725 [2:49:07<4:42:01,  1.83s/it]
 37%|███▋      | 5482/14725 [2:49:08<4:41:21,  1.83s/it]
 37%|███▋      | 5483/14725 [2:49:10<4:40:43,  1.82s/it]
 37%|███▋      | 5484/14725 [2:49:12<4:38:00,  1.81s/it]
 37%|███▋      | 5485/14725 [2:49:14<4:38:24,  1.81s/it]
 37%|███▋      | 5486/14725 [2:49:16<4:41:40,  1.83s/it]
 37%|███▋      | 5487/14725 [2:49:17<4:40:48,  1.82s/it]
 37%|███▋      | 5488/14725 [2:49:19<4:41:15,  1.83s/it]
 37%|�
+0: {'loss': 0.8409, 'grad_norm': 0.6761011757511645, 'learning_rate': 2e-05, 'epoch': 0.37}
+0: {'loss': 0.8266, 'grad_norm': 0.7638129774503376, 'learning_rate': 2e-05, 'epoch': 0.37}
+0: ��██▋      | 5489/14725 [2:49:21<4:40:53,  1.82s/it]
 37%|███▋      | 5490/14725 [2:49:23<4:40:16,  1.82s/it]
                                                        

 37%|███▋      | 5490/14725 [2:49:23<4:40:16,  1.82s/it]
 37%|███▋      | 5491/14725 [2:49:25<4:42:25,  1.84s/it]
 37%|███▋      | 5492/14725 [2:49:27<4:39:33,  1.82s/it]
 37%|███▋      | 5493/14725 [2:49:28<4:40:33,  1.82s/it]
 37%|███▋      | 5494/14725 [2:49:30<4:38:34,  1.81s/it]
 37%|███▋      | 5495/14725 [2:49:32<4:36:57,  1.80s/it]
 37%|███▋      | 5496/14725 [2:49:34<4:36:38,  1.80s/it]
 37%|███▋      | 5497/14725 [2:49:36<4:37:36,  1.81s/it]
 37%|███▋      | 5498/14725 [2:49:37<4:38:46,  1.81s/it]
 37%|███▋      | 5499/14725 [2:49:39<4:37:49,  1.81s/it]
 37%|███▋      | 5500/14725 [2:49:41<4:35:32,  1.79s/it]
                                                        

 37%|███▋      | 5500/14725 [2:49:41<4:35:32,  1.79s/it]
 37%
+0: {'loss': 0.8338, 'grad_norm': 0.7642046969153675, 'learning_rate': 2e-05, 'epoch': 0.37}
+0: |███▋      | 5501/14725 [2:49:43<4:33:58,  1.78s/it]
 37%|███▋      | 5502/14725 [2:49:44<4:32:35,  1.77s/it]
 37%|███▋      | 5503/14725 [2:49:46<4:32:39,  1.77s/it]
 37%|███▋      | 5504/14725 [2:49:48<4:31:31,  1.77s/it]
 37%|███▋      | 5505/14725 [2:49:50<4:31:54,  1.77s/it]
 37%|███▋      | 5506/14725 [2:49:52<4:31:25,  1.77s/it]
 37%|███▋      | 5507/14725 [2:49:53<4:32:59,  1.78s/it]
 37%|███▋      | 5508/14725 [2:49:55<4:34:24,  1.79s/it]
 37%|███▋      | 5509/14725 [2:49:57<4:35:01,  1.79s/it]
 37%|███▋      | 5510/14725 [2:49:59<4:33:18,  1.78s/it]
                                                        

 37%|███▋      | 5510/14725 [2:49:59<4:33:18,  1.78s/it]
 37%|███▋      | 5511/14725 [2:50:01<4:36:06,  1.80s/it]
 37%|███▋      | 5512/14725 [2:50:02<4:34:48,  1.79s/it]
 37%|███▋      | 5513/14725 [2:50:04<4:35:46,  1.80s/it]
 37%|███▋      | 5514/14725 [2:50:06<4:33:58,  1.78s
+0: {'loss': 0.8407, 'grad_norm': 0.796270766465346, 'learning_rate': 2e-05, 'epoch': 0.37}
+0: /it]
 37%|███▋      | 5515/14725 [2:50:08<4:38:49,  1.82s/it]
 37%|███▋      | 5516/14725 [2:50:10<4:40:23,  1.83s/it]
 37%|███▋      | 5517/14725 [2:50:11<4:40:35,  1.83s/it]
 37%|███▋      | 5518/14725 [2:50:13<4:37:01,  1.81s/it]
 37%|███▋      | 5519/14725 [2:50:15<4:33:55,  1.79s/it]
 37%|███▋      | 5520/14725 [2:50:17<4:35:06,  1.79s/it]
                                                        

 37%|███▋      | 5520/14725 [2:50:17<4:35:06,  1.79s/it]
 37%|███▋      | 5521/14725 [2:50:19<4:36:12,  1.80s/it]
 38%|███▊      | 5522/14725 [2:50:20<4:36:34,  1.80s/it]
 38%|███▊      | 5523/14725 [2:50:22<4:35:20,  1.80s/it]
 38%|███▊      | 5524/14725 [2:50:24<4:33:19,  1.78s/it]
 38%|███▊      | 5525/14725 [2:50:26<4:33:29,  1.78s/it]
 38%|███▊      | 5526/14725 [2:50:28<4:39:03,  1.82s/it]
 38%|███▊      | 5527/14725 [2:50:29<4:38:11,  1.81s/it]
 38%|███▊      | 5528/14725 [2:50:31<4:36:2
+0: {'loss': 0.8371, 'grad_norm': 0.8638901805991193, 'learning_rate': 2e-05, 'epoch': 0.38}
+0: {'loss': 0.8427, 'grad_norm': 0.7393987934575653, 'learning_rate': 2e-05, 'epoch': 0.38}
+0: 2,  1.80s/it]
 38%|███▊      | 5529/14725 [2:50:33<4:38:40,  1.82s/it]
 38%|███▊      | 5530/14725 [2:50:35<4:43:42,  1.85s/it]
                                                        

 38%|███▊      | 5530/14725 [2:50:35<4:43:42,  1.85s/it]
 38%|███▊      | 5531/14725 [2:50:37<4:42:12,  1.84s/it]
 38%|███▊      | 5532/14725 [2:50:39<4:41:33,  1.84s/it]
 38%|███▊      | 5533/14725 [2:50:40<4:39:47,  1.83s/it]
 38%|███▊      | 5534/14725 [2:50:42<4:37:42,  1.81s/it]
 38%|███▊      | 5535/14725 [2:50:44<4:34:52,  1.79s/it]
 38%|███▊      | 5536/14725 [2:50:46<4:36:14,  1.80s/it]
 38%|███▊      | 5537/14725 [2:50:48<4:35:53,  1.80s/it]
 38%|███▊      | 5538/14725 [2:50:49<4:38:15,  1.82s/it]
 38%|███▊      | 5539/14725 [2:50:51<4:40:32,  1.83s/it]
 38%|███▊      | 5540/14725 [2:50:53<4:40:06,  1.83s/it]
                                                        

 38%|███▊      | 5540/14725 [2:50:53<4:40
+0: {'loss': 0.8276, 'grad_norm': 0.7098856750961663, 'learning_rate': 2e-05, 'epoch': 0.38}
+0: :06,  1.83s/it]
 38%|███▊      | 5541/14725 [2:50:55<4:42:52,  1.85s/it]
 38%|███▊      | 5542/14725 [2:50:57<4:45:12,  1.86s/it]
 38%|███▊      | 5543/14725 [2:50:59<4:43:10,  1.85s/it]
 38%|███▊      | 5544/14725 [2:51:00<4:40:44,  1.83s/it]
 38%|███▊      | 5545/14725 [2:51:02<4:36:21,  1.81s/it]
 38%|███▊      | 5546/14725 [2:51:04<4:34:28,  1.79s/it]
 38%|███▊      | 5547/14725 [2:51:06<4:31:53,  1.78s/it]
 38%|███▊      | 5548/14725 [2:51:08<4:33:43,  1.79s/it]
 38%|███▊      | 5549/14725 [2:51:09<4:32:05,  1.78s/it]
 38%|███▊      | 5550/14725 [2:51:11<4:32:21,  1.78s/it]
                                                        

 38%|███▊      | 5550/14725 [2:51:11<4:32:21,  1.78s/it]
 38%|███▊      | 5551/14725 [2:51:13<4:39:17,  1.83s/it]
 38%|███▊      | 5552/14725 [2:51:15<4:37:34,  1.82s/it]
 38%|███▊      | 5553/14725 [2:51:17<4:34:41,  1.80s/it]
 38%|███▊      | 5554/14725 [2:5
+0: {'loss': 0.8376, 'grad_norm': 0.7416424282483993, 'learning_rate': 2e-05, 'epoch': 0.38}
+0: 1:18<4:35:23,  1.80s/it]
 38%|███▊      | 5555/14725 [2:51:20<4:35:27,  1.80s/it]
 38%|███▊      | 5556/14725 [2:51:22<4:32:19,  1.78s/it]
 38%|███▊      | 5557/14725 [2:51:24<4:31:12,  1.77s/it]
 38%|███▊      | 5558/14725 [2:51:25<4:32:36,  1.78s/it]
 38%|███▊      | 5559/14725 [2:51:27<4:30:51,  1.77s/it]
 38%|███▊      | 5560/14725 [2:51:29<4:29:59,  1.77s/it]
                                                        

 38%|███▊      | 5560/14725 [2:51:29<4:29:59,  1.77s/it]
 38%|███▊      | 5561/14725 [2:51:31<4:31:47,  1.78s/it]
 38%|███▊      | 5562/14725 [2:51:33<4:31:01,  1.77s/it]
 38%|███▊      | 5563/14725 [2:51:34<4:33:56,  1.79s/it]
 38%|███▊      | 5564/14725 [2:51:36<4:34:46,  1.80s/it]
 38%|███▊      | 5565/14725 [2:51:38<4:36:13,  1.81s/it]
 38%|███▊      | 5566/14725 [2:51:40<4:38:31,  1.82s/it]
 38%|███▊      | 5567/14725 [2:51:42<4:39:27,  1.83s/it]
 38%|███▊      | 5568/1
+0: {'loss': 0.8482, 'grad_norm': 0.7535867055113257, 'learning_rate': 2e-05, 'epoch': 0.38}
+0: {'loss': 0.8398, 'grad_norm': 0.8199838471183569, 'learning_rate': 2e-05, 'epoch': 0.38}
+0: 4725 [2:51:44<4:38:15,  1.82s/it]
 38%|███▊      | 5569/14725 [2:51:45<4:35:06,  1.80s/it]
 38%|███▊      | 5570/14725 [2:51:47<4:37:14,  1.82s/it]
                                                        

 38%|███▊      | 5570/14725 [2:51:47<4:37:14,  1.82s/it]
 38%|███▊      | 5571/14725 [2:51:49<4:40:38,  1.84s/it]
 38%|███▊      | 5572/14725 [2:51:51<4:39:35,  1.83s/it]
 38%|███▊      | 5573/14725 [2:51:53<4:37:51,  1.82s/it]
 38%|███▊      | 5574/14725 [2:51:55<4:40:52,  1.84s/it]
 38%|███▊      | 5575/14725 [2:51:56<4:37:48,  1.82s/it]
 38%|███▊      | 5576/14725 [2:51:58<4:34:43,  1.80s/it]
 38%|███▊      | 5577/14725 [2:52:00<4:33:27,  1.79s/it]
 38%|███▊      | 5578/14725 [2:52:02<4:36:58,  1.82s/it]
 38%|███▊      | 5579/14725 [2:52:04<4:35:31,  1.81s/it]
 38%|███▊      | 5580/14725 [2:52:05<4:36:14,  1.81s/it]
                                                        

 38%|███▊      | 5580
+0: {'loss': 0.8372, 'grad_norm': 0.7507705341607553, 'learning_rate': 2e-05, 'epoch': 0.38}
+0: /14725 [2:52:05<4:36:14,  1.81s/it]
 38%|███▊      | 5581/14725 [2:52:07<4:34:19,  1.80s/it]
 38%|███▊      | 5582/14725 [2:52:09<4:35:09,  1.81s/it]
 38%|███▊      | 5583/14725 [2:52:11<4:35:54,  1.81s/it]
 38%|███▊      | 5584/14725 [2:52:13<4:35:08,  1.81s/it]
 38%|███▊      | 5585/14725 [2:52:14<4:35:34,  1.81s/it]
 38%|███▊      | 5586/14725 [2:52:16<4:36:52,  1.82s/it]
 38%|███▊      | 5587/14725 [2:52:18<4:37:11,  1.82s/it]
 38%|███▊      | 5588/14725 [2:52:20<4:40:28,  1.84s/it]
 38%|███▊      | 5589/14725 [2:52:22<4:36:49,  1.82s/it]
 38%|███▊      | 5590/14725 [2:52:23<4:36:12,  1.81s/it]
                                                        

 38%|███▊      | 5590/14725 [2:52:23<4:36:12,  1.81s/it]
 38%|███▊      | 5591/14725 [2:52:25<4:35:20,  1.81s/it]
 38%|███▊      | 5592/14725 [2:52:27<4:34:08,  1.80s/it]
 38%|███▊      | 5593/14725 [2:52:29<4:33:33,  1.80s/it]
 38%|███▊   
+0: {'loss': 0.8448, 'grad_norm': 0.7978076923896681, 'learning_rate': 2e-05, 'epoch': 0.38}
+0:    | 5594/14725 [2:52:31<4:35:03,  1.81s/it]
 38%|███▊      | 5595/14725 [2:52:32<4:35:10,  1.81s/it]
 38%|███▊      | 5596/14725 [2:52:34<4:34:54,  1.81s/it]
 38%|███▊      | 5597/14725 [2:52:36<4:32:55,  1.79s/it]
 38%|███▊      | 5598/14725 [2:52:38<4:37:23,  1.82s/it]
 38%|███▊      | 5599/14725 [2:52:40<4:39:42,  1.84s/it]
 38%|███▊      | 5600/14725 [2:52:42<4:35:30,  1.81s/it]
                                                        

 38%|███▊      | 5600/14725 [2:52:42<4:35:30,  1.81s/it]
 38%|███▊      | 5601/14725 [2:52:43<4:35:51,  1.81s/it]
 38%|███▊      | 5602/14725 [2:52:45<4:36:40,  1.82s/it]
 38%|███▊      | 5603/14725 [2:52:47<4:33:49,  1.80s/it]
 38%|███▊      | 5604/14725 [2:52:49<4:32:59,  1.80s/it]
 38%|███▊      | 5605/14725 [2:52:51<4:32:05,  1.79s/it]
 38%|███▊      | 5606/14725 [2:52:52<4:29:46,  1.78s/it]
 38%|███▊      | 5607/14725 [2:52:54<4:30:29,  1.78s/it]
 38%|██
+0: {'loss': 0.8406, 'grad_norm': 0.7453718899671472, 'learning_rate': 2e-05, 'epoch': 0.38}
+0: {'loss': 0.8391, 'grad_norm': 0.734292563042069, 'learning_rate': 2e-05, 'epoch': 0.38}
+0: █▊      | 5608/14725 [2:52:56<4:30:23,  1.78s/it]
 38%|███▊      | 5609/14725 [2:52:58<4:31:45,  1.79s/it]
 38%|███▊      | 5610/14725 [2:52:59<4:30:56,  1.78s/it]
                                                        

 38%|███▊      | 5610/14725 [2:52:59<4:30:56,  1.78s/it]
 38%|███▊      | 5611/14725 [2:53:01<4:30:44,  1.78s/it]
 38%|███▊      | 5612/14725 [2:53:03<4:43:35,  1.87s/it]
 38%|███▊      | 5613/14725 [2:53:05<4:40:21,  1.85s/it]
 38%|███▊      | 5614/14725 [2:53:07<4:42:26,  1.86s/it]
 38%|███▊      | 5615/14725 [2:53:09<4:37:50,  1.83s/it]
 38%|███▊      | 5616/14725 [2:53:11<4:36:43,  1.82s/it]
 38%|███▊      | 5617/14725 [2:53:12<4:34:16,  1.81s/it]
 38%|███▊      | 5618/14725 [2:53:14<4:33:17,  1.80s/it]
 38%|███▊      | 5619/14725 [2:53:16<4:39:22,  1.84s/it]
 38%|███▊      | 5620/14725 [2:53:18<4:37:32,  1.83s/it]
                                                        

 38%|█�
+0: {'loss': 0.8402, 'grad_norm': 0.7873354253764304, 'learning_rate': 2e-05, 'epoch': 0.38}
+0: ��█▊      | 5620/14725 [2:53:18<4:37:32,  1.83s/it]
 38%|███▊      | 5621/14725 [2:53:20<4:37:59,  1.83s/it]
 38%|███▊      | 5622/14725 [2:53:22<4:38:28,  1.84s/it]
 38%|███▊      | 5623/14725 [2:53:23<4:38:17,  1.83s/it]
 38%|███▊      | 5624/14725 [2:53:25<4:34:16,  1.81s/it]
 38%|███▊      | 5625/14725 [2:53:27<4:34:34,  1.81s/it]
 38%|███▊      | 5626/14725 [2:53:29<4:36:33,  1.82s/it]
 38%|███▊      | 5627/14725 [2:53:31<4:33:12,  1.80s/it]
 38%|███▊      | 5628/14725 [2:53:32<4:31:13,  1.79s/it]
 38%|███▊      | 5629/14725 [2:53:34<4:32:20,  1.80s/it]
 38%|███▊      | 5630/14725 [2:53:36<4:36:12,  1.82s/it]
                                                        

 38%|███▊      | 5630/14725 [2:53:36<4:36:12,  1.82s/it]
 38%|███▊      | 5631/14725 [2:53:38<4:35:24,  1.82s/it]
 38%|███▊      | 5632/14725 [2:53:40<4:34:14,  1.81s/it]
 38%|███▊      | 5633/14725 [2:53:41<4:34:42,  1.81s/it]
+0: {'loss': 0.8551, 'grad_norm': 0.7437331220648615, 'learning_rate': 2e-05, 'epoch': 0.38}
+0:  38%|███▊      | 5634/14725 [2:53:43<4:31:03,  1.79s/it]
 38%|███▊      | 5635/14725 [2:53:45<4:29:15,  1.78s/it]
 38%|███▊      | 5636/14725 [2:53:47<4:35:23,  1.82s/it]
 38%|███▊      | 5637/14725 [2:53:49<4:33:57,  1.81s/it]
 38%|███▊      | 5638/14725 [2:53:50<4:32:47,  1.80s/it]
 38%|███▊      | 5639/14725 [2:53:52<4:31:25,  1.79s/it]
 38%|███▊      | 5640/14725 [2:53:54<4:33:58,  1.81s/it]
                                                        

 38%|███▊      | 5640/14725 [2:53:54<4:33:58,  1.81s/it]
 38%|███▊      | 5641/14725 [2:53:56<4:33:47,  1.81s/it]
 38%|███▊      | 5642/14725 [2:53:58<4:33:10,  1.80s/it]
 38%|███▊      | 5643/14725 [2:54:00<4:39:55,  1.85s/it]
 38%|███▊      | 5644/14725 [2:54:01<4:38:30,  1.84s/it]
 38%|███▊      | 5645/14725 [2:54:03<4:40:30,  1.85s/it]
 38%|███▊      | 5646/14725 [2:54:05<4:38:00,  1.84s/it]
 38%|███▊      | 5647/14725 [2:54:07<4:37:10,  1
+0: {'loss': 0.8343, 'grad_norm': 0.7370602752952269, 'learning_rate': 2e-05, 'epoch': 0.38}
+0: .83s/it]
 38%|███▊      | 5648/14725 [2:54:09<4:33:23,  1.81s/it]
 38%|███▊      | 5649/14725 [2:54:10<4:35:39,  1.82s/it]
 38%|███▊      | 5650/14725 [2:54:12<4:37:53,  1.84s/it]
                                                        

 38%|███▊      | 5650/14725 [2:54:12<4:37:53,  1.84s/it]
 38%|███▊      | 5651/14725 [2:54:14<4:38:04,  1.84s/it]
 38%|███▊      | 5652/14725 [2:54:16<4:34:51,  1.82s/it]
 38%|███▊      | 5653/14725 [2:54:18<4:31:40,  1.80s/it]
 38%|███▊      | 5654/14725 [2:54:19<4:31:28,  1.80s/it]
 38%|███▊      | 5655/14725 [2:54:21<4:33:31,  1.81s/it]
 38%|███▊      | 5656/14725 [2:54:23<4:31:33,  1.80s/it]
 38%|███▊      | 5657/14725 [2:54:25<4:33:01,  1.81s/it]
 38%|███▊      | 5658/14725 [2:54:27<4:31:51,  1.80s/it]
 38%|███▊      | 5659/14725 [2:54:29<4:32:27,  1.80s/it]
 38%|███▊      | 5660/14725 [2:54:30<4:30:52,  1.79s/it]
                                               
+0: {'loss': 0.8288, 'grad_norm': 0.71618579483591, 'learning_rate': 2e-05, 'epoch': 0.38}
+0: {'loss': 0.8298, 'grad_norm': 0.7720033200169754, 'learning_rate': 2e-05, 'epoch': 0.39}
+0:          

 38%|███▊      | 5660/14725 [2:54:30<4:30:52,  1.79s/it]
 38%|███▊      | 5661/14725 [2:54:32<4:34:53,  1.82s/it]
 38%|███▊      | 5662/14725 [2:54:34<4:31:18,  1.80s/it]
 38%|███▊      | 5663/14725 [2:54:36<4:30:56,  1.79s/it]
 38%|███▊      | 5664/14725 [2:54:37<4:29:32,  1.78s/it]
 38%|███▊      | 5665/14725 [2:54:39<4:27:25,  1.77s/it]
 38%|███▊      | 5666/14725 [2:54:41<4:27:44,  1.77s/it]
 38%|███▊      | 5667/14725 [2:54:43<4:29:08,  1.78s/it]
 38%|███▊      | 5668/14725 [2:54:45<4:31:44,  1.80s/it]
 38%|███▊      | 5669/14725 [2:54:46<4:31:12,  1.80s/it]
 39%|███▊      | 5670/14725 [2:54:48<4:31:32,  1.80s/it]
                                                        

 39%|███▊      | 5670/14725 [2:54:48<4:31:32,  1.80s/it]
 39%|███▊      | 5671/14725 [2:54:50<4:32:27,  1.81s/it]
 39%|███▊      | 5672/14725 [2:54:52<4:34:00,  1.82s/it]
 39%|███▊      | 5673/14725 [2:54:54<
+0: {'loss': 0.8538, 'grad_norm': 0.7338086103154653, 'learning_rate': 2e-05, 'epoch': 0.39}
+0: 4:37:45,  1.84s/it]
 39%|███▊      | 5674/14725 [2:54:56<4:35:11,  1.82s/it]
 39%|███▊      | 5675/14725 [2:54:57<4:33:47,  1.82s/it]
 39%|███▊      | 5676/14725 [2:54:59<4:33:34,  1.81s/it]
 39%|███▊      | 5677/14725 [2:55:01<4:33:19,  1.81s/it]
 39%|███▊      | 5678/14725 [2:55:03<4:30:13,  1.79s/it]
 39%|███▊      | 5679/14725 [2:55:04<4:29:40,  1.79s/it]
 39%|███▊      | 5680/14725 [2:55:06<4:33:54,  1.82s/it]
                                                        

 39%|███▊      | 5680/14725 [2:55:06<4:33:54,  1.82s/it]
 39%|███▊      | 5681/14725 [2:55:08<4:34:55,  1.82s/it]
 39%|███▊      | 5682/14725 [2:55:10<4:33:57,  1.82s/it]
 39%|███▊      | 5683/14725 [2:55:12<4:34:03,  1.82s/it]
 39%|███▊      | 5684/14725 [2:55:14<4:32:48,  1.81s/it]
 39%|███▊      | 5685/14725 [2:55:15<4:34:56,  1.82s/it]
 39%|███▊      | 5686/14725 [2:55:17<4:32:25,  1.81s/it]
 39%|███▊      | 5687/14725 
+0: {'loss': 0.8312, 'grad_norm': 0.7275976279809958, 'learning_rate': 2e-05, 'epoch': 0.39}
+0: [2:55:19<4:33:21,  1.81s/it]
 39%|███▊      | 5688/14725 [2:55:21<4:32:38,  1.81s/it]
 39%|███▊      | 5689/14725 [2:55:23<4:32:03,  1.81s/it]
 39%|███▊      | 5690/14725 [2:55:24<4:31:02,  1.80s/it]
                                                        

 39%|███▊      | 5690/14725 [2:55:24<4:31:02,  1.80s/it]
 39%|███▊      | 5691/14725 [2:55:26<4:30:22,  1.80s/it]
 39%|███▊      | 5692/14725 [2:55:28<4:30:11,  1.79s/it]
 39%|███▊      | 5693/14725 [2:55:30<4:31:43,  1.81s/it]
 39%|███▊      | 5694/14725 [2:55:32<4:30:33,  1.80s/it]
 39%|███▊      | 5695/14725 [2:55:33<4:30:21,  1.80s/it]
 39%|███▊      | 5696/14725 [2:55:35<4:28:17,  1.78s/it]
 39%|███▊      | 5697/14725 [2:55:37<4:26:44,  1.77s/it]
 39%|███▊      | 5698/14725 [2:55:39<4:28:09,  1.78s/it]
 39%|███▊      | 5699/14725 [2:55:40<4:26:01,  1.77s/it]
 39%|███▊      | 5700/14725 [2:55:42<4:29:03,  1.79s/it]
                           
+0: {'loss': 0.8363, 'grad_norm': 0.7312006645389677, 'learning_rate': 2e-05, 'epoch': 0.39}
+0: {'loss': 0.8256, 'grad_norm': 0.7352816997846581, 'learning_rate': 2e-05, 'epoch': 0.39}
+0:                              

 39%|███▊      | 5700/14725 [2:55:42<4:29:03,  1.79s/it]
 39%|███▊      | 5701/14725 [2:55:44<4:28:12,  1.78s/it]
 39%|███▊      | 5702/14725 [2:55:46<4:27:17,  1.78s/it]
 39%|███▊      | 5703/14725 [2:55:48<4:26:30,  1.77s/it]
 39%|███▊      | 5704/14725 [2:55:49<4:25:53,  1.77s/it]
 39%|███▊      | 5705/14725 [2:55:51<4:28:14,  1.78s/it]
 39%|███▉      | 5706/14725 [2:55:53<4:29:45,  1.79s/it]
 39%|███▉      | 5707/14725 [2:55:55<4:31:16,  1.80s/it]
 39%|███▉      | 5708/14725 [2:55:57<4:29:54,  1.80s/it]
 39%|███▉      | 5709/14725 [2:55:58<4:29:34,  1.79s/it]
 39%|███▉      | 5710/14725 [2:56:00<4:28:39,  1.79s/it]
                                                        

 39%|███▉      | 5710/14725 [2:56:00<4:28:39,  1.79s/it]
 39%|███▉      | 5711/14725 [2:56:02<4:30:19,  1.80s/it]
 39%|███▉      | 5712/14725 [2:56:04<4:30:32,  1.80s/it]
 39%|███▉      | 
+0: {'loss': 0.8263, 'grad_norm': 0.7431638474931321, 'learning_rate': 2e-05, 'epoch': 0.39}
+0: 5713/14725 [2:56:06<4:28:06,  1.78s/it]
 39%|███▉      | 5714/14725 [2:56:07<4:30:16,  1.80s/it]
 39%|███▉      | 5715/14725 [2:56:09<4:29:19,  1.79s/it]
 39%|███▉      | 5716/14725 [2:56:11<4:27:39,  1.78s/it]
 39%|███▉      | 5717/14725 [2:56:13<4:28:46,  1.79s/it]
 39%|███▉      | 5718/14725 [2:56:15<4:29:56,  1.80s/it]
 39%|███▉      | 5719/14725 [2:56:16<4:27:28,  1.78s/it]
 39%|███▉      | 5720/14725 [2:56:18<4:27:16,  1.78s/it]
                                                        

 39%|███▉      | 5720/14725 [2:56:18<4:27:16,  1.78s/it]
 39%|███▉      | 5721/14725 [2:56:20<4:27:30,  1.78s/it]
 39%|███▉      | 5722/14725 [2:56:22<4:26:44,  1.78s/it]
 39%|███▉      | 5723/14725 [2:56:23<4:29:59,  1.80s/it]
 39%|███▉      | 5724/14725 [2:56:25<4:30:55,  1.81s/it]
 39%|███▉      | 5725/14725 [2:56:27<4:29:05,  1.79s/it]
 39%|███▉      | 5726/14725 [2:56:29<4:27:56,  1.79s/it]
 39%|███�
+0: {'loss': 0.8238, 'grad_norm': 0.7907732905363674, 'learning_rate': 2e-05, 'epoch': 0.39}
+0: �      | 5727/14725 [2:56:31<4:26:28,  1.78s/it]
 39%|███▉      | 5728/14725 [2:56:32<4:24:57,  1.77s/it]
 39%|███▉      | 5729/14725 [2:56:34<4:24:40,  1.77s/it]
 39%|███▉      | 5730/14725 [2:56:36<4:27:31,  1.78s/it]
                                                        

 39%|███▉      | 5730/14725 [2:56:36<4:27:31,  1.78s/it]
 39%|███▉      | 5731/14725 [2:56:38<4:28:48,  1.79s/it]
 39%|███▉      | 5732/14725 [2:56:40<4:30:32,  1.81s/it]
 39%|███▉      | 5733/14725 [2:56:41<4:32:36,  1.82s/it]
 39%|███▉      | 5734/14725 [2:56:43<4:33:23,  1.82s/it]
 39%|███▉      | 5735/14725 [2:56:45<4:32:09,  1.82s/it]
 39%|███▉      | 5736/14725 [2:56:47<4:32:22,  1.82s/it]
 39%|███▉      | 5737/14725 [2:56:49<4:31:08,  1.81s/it]
 39%|███▉      | 5738/14725 [2:56:50<4:28:52,  1.80s/it]
 39%|███▉      | 5739/14725 [2:56:52<4:27:31,  1.79s/it]
 39%|███▉      | 5740/14725 [2:56:54<4:30:22,  1.81s/it]
       
+0: {'loss': 0.8258, 'grad_norm': 0.7568158043441318, 'learning_rate': 2e-05, 'epoch': 0.39}
+0: {'loss': 0.8444, 'grad_norm': 0.760481766718096, 'learning_rate': 2e-05, 'epoch': 0.39}
+0:                                                  

 39%|███▉      | 5740/14725 [2:56:54<4:30:22,  1.81s/it]
 39%|███▉      | 5741/14725 [2:56:56<4:30:43,  1.81s/it]
 39%|███▉      | 5742/14725 [2:56:58<4:29:10,  1.80s/it]
 39%|███▉      | 5743/14725 [2:56:59<4:29:35,  1.80s/it]
 39%|███▉      | 5744/14725 [2:57:01<4:28:31,  1.79s/it]
 39%|███▉      | 5745/14725 [2:57:03<4:32:36,  1.82s/it]
 39%|███▉      | 5746/14725 [2:57:05<4:30:39,  1.81s/it]
 39%|███▉      | 5747/14725 [2:57:07<4:34:11,  1.83s/it]
 39%|███▉      | 5748/14725 [2:57:09<4:30:20,  1.81s/it]
 39%|███▉      | 5749/14725 [2:57:10<4:32:56,  1.82s/it]
 39%|███▉      | 5750/14725 [2:57:12<4:33:52,  1.83s/it]
                                                        

 39%|███▉      | 5750/14725 [2:57:12<4:33:52,  1.83s/it]
 39%|███▉      | 5751/14725 [2:57:14<4:31:24,  1.81s/it]
 39%|███▉      | 5752/14725 [2:57:16<4:34:39,  1.84s/it]
 39%|
+0: {'loss': 0.8279, 'grad_norm': 0.7160917404589868, 'learning_rate': 2e-05, 'epoch': 0.39}
+0: ███▉      | 5753/14725 [2:57:18<4:31:36,  1.82s/it]
 39%|███▉      | 5754/14725 [2:57:20<4:32:17,  1.82s/it]
 39%|███▉      | 5755/14725 [2:57:21<4:36:25,  1.85s/it]
 39%|███▉      | 5756/14725 [2:57:23<4:32:34,  1.82s/it]
 39%|███▉      | 5757/14725 [2:57:25<4:29:04,  1.80s/it]
 39%|███▉      | 5758/14725 [2:57:27<4:32:46,  1.83s/it]
 39%|███▉      | 5759/14725 [2:57:29<4:34:09,  1.83s/it]
 39%|███▉      | 5760/14725 [2:57:30<4:31:43,  1.82s/it]
                                                        

 39%|███▉      | 5760/14725 [2:57:30<4:31:43,  1.82s/it]
 39%|███▉      | 5761/14725 [2:57:32<4:32:54,  1.83s/it]
 39%|███▉      | 5762/14725 [2:57:34<4:33:05,  1.83s/it]
 39%|███▉      | 5763/14725 [2:57:36<4:34:51,  1.84s/it]
 39%|███▉      | 5764/14725 [2:57:38<4:33:24,  1.83s/it]
 39%|███▉      | 5765/14725 [2:57:40<4:30:28,  1.81s/it]
 39%|███▉      | 5766/14725 [2:57:41<4:30:30,  1.81s/
+0: {'loss': 0.8206, 'grad_norm': 0.7731439794112457, 'learning_rate': 2e-05, 'epoch': 0.39}
+0: it]
 39%|███▉      | 5767/14725 [2:57:43<4:31:00,  1.82s/it]
 39%|███▉      | 5768/14725 [2:57:45<4:29:21,  1.80s/it]
 39%|███▉      | 5769/14725 [2:57:47<4:28:52,  1.80s/it]
 39%|███▉      | 5770/14725 [2:57:49<4:30:03,  1.81s/it]
                                                        

 39%|███▉      | 5770/14725 [2:57:49<4:30:03,  1.81s/it]
 39%|███▉      | 5771/14725 [2:57:50<4:30:49,  1.81s/it]
 39%|███▉      | 5772/14725 [2:57:52<4:28:53,  1.80s/it]
 39%|███▉      | 5773/14725 [2:57:54<4:28:24,  1.80s/it]
 39%|███▉      | 5774/14725 [2:57:56<4:28:49,  1.80s/it]
 39%|███▉      | 5775/14725 [2:57:58<4:30:25,  1.81s/it]
 39%|███▉      | 5776/14725 [2:57:59<4:28:41,  1.80s/it]
 39%|███▉      | 5777/14725 [2:58:01<4:27:44,  1.80s/it]
 39%|███▉      | 5778/14725 [2:58:03<4:27:27,  1.79s/it]
 39%|███▉      | 5779/14725 [2:58:05<4:31:42,  1.82s/it]
 39%|███▉      | 5780/14725 [2:58:07<4:28:27
+0: {'loss': 0.8459, 'grad_norm': 0.9339772331914887, 'learning_rate': 2e-05, 'epoch': 0.39}
+0: {'loss': 0.8306, 'grad_norm': 0.6982594169955963, 'learning_rate': 2e-05, 'epoch': 0.39}
+0: ,  1.80s/it]
                                                        

 39%|███▉      | 5780/14725 [2:58:07<4:28:27,  1.80s/it]
 39%|███▉      | 5781/14725 [2:58:09<4:33:49,  1.84s/it]
 39%|███▉      | 5782/14725 [2:58:10<4:33:30,  1.83s/it]
 39%|███▉      | 5783/14725 [2:58:12<4:31:21,  1.82s/it]
 39%|███▉      | 5784/14725 [2:58:14<4:29:52,  1.81s/it]
 39%|███▉      | 5785/14725 [2:58:16<4:29:10,  1.81s/it]
 39%|███▉      | 5786/14725 [2:58:18<4:27:37,  1.80s/it]
 39%|███▉      | 5787/14725 [2:58:19<4:31:46,  1.82s/it]
 39%|███▉      | 5788/14725 [2:58:21<4:30:05,  1.81s/it]
 39%|███▉      | 5789/14725 [2:58:23<4:27:00,  1.79s/it]
 39%|███▉      | 5790/14725 [2:58:25<4:27:58,  1.80s/it]
                                                        

 39%|███▉      | 5790/14725 [2:58:25<4:27:58,  1.80s/it]
 39%|███▉      | 5791/14725 [2:58:27<4:28:46,  1.81s/it]
 39%|███▉      | 5792/14725 [2:58:28<4:27:
+0: {'loss': 0.8341, 'grad_norm': 0.694260526622164, 'learning_rate': 2e-05, 'epoch': 0.39}
+0: 02,  1.79s/it]
 39%|███▉      | 5793/14725 [2:58:30<4:28:05,  1.80s/it]
 39%|███▉      | 5794/14725 [2:58:32<4:26:02,  1.79s/it]
 39%|███▉      | 5795/14725 [2:58:34<4:25:41,  1.79s/it]
 39%|███▉      | 5796/14725 [2:58:36<4:30:10,  1.82s/it]
 39%|███▉      | 5797/14725 [2:58:37<4:29:19,  1.81s/it]
 39%|███▉      | 5798/14725 [2:58:39<4:27:35,  1.80s/it]
 39%|███▉      | 5799/14725 [2:58:41<4:30:00,  1.81s/it]
 39%|███▉      | 5800/14725 [2:58:43<4:29:31,  1.81s/it]
                                                        

 39%|███▉      | 5800/14725 [2:58:43<4:29:31,  1.81s/it]
 39%|███▉      | 5801/14725 [2:58:45<4:28:36,  1.81s/it]
 39%|███▉      | 5802/14725 [2:58:46<4:27:31,  1.80s/it]
 39%|███▉      | 5803/14725 [2:58:48<4:34:06,  1.84s/it]
 39%|███▉      | 5804/14725 [2:58:50<4:33:31,  1.84s/it]
 39%|███▉      | 5805/14725 [2:58:52<4:35:00,  1.85s/it]
 39%|███▉      | 5806/14725 [2:58
+0: {'loss': 0.8525, 'grad_norm': 0.741931108289796, 'learning_rate': 2e-05, 'epoch': 0.39}
+0: :54<4:34:45,  1.85s/it]
 39%|███▉      | 5807/14725 [2:58:56<4:35:32,  1.85s/it]
 39%|███▉      | 5808/14725 [2:58:58<4:40:36,  1.89s/it]
 39%|███▉      | 5809/14725 [2:58:59<4:34:40,  1.85s/it]
 39%|███▉      | 5810/14725 [2:59:01<4:32:23,  1.83s/it]
                                                        

 39%|███▉      | 5810/14725 [2:59:01<4:32:23,  1.83s/it]
 39%|███▉      | 5811/14725 [2:59:03<4:32:01,  1.83s/it]
 39%|███▉      | 5812/14725 [2:59:05<4:33:39,  1.84s/it]
 39%|███▉      | 5813/14725 [2:59:07<4:33:25,  1.84s/it]
 39%|███▉      | 5814/14725 [2:59:09<4:31:37,  1.83s/it]
 39%|███▉      | 5815/14725 [2:59:10<4:29:03,  1.81s/it]
 39%|███▉      | 5816/14725 [2:59:12<4:29:51,  1.82s/it]
 40%|███▉      | 5817/14725 [2:59:14<4:27:27,  1.80s/it]
 40%|███▉      | 5818/14725 [2:59:16<4:25:04,  1.79s/it]
 40%|███▉      | 5819/14725 [2:59:18<4:24:23,  1.78s/it]
 40%|███▉      | 5820/14
+0: {'loss': 0.8454, 'grad_norm': 0.7784437109280148, 'learning_rate': 2e-05, 'epoch': 0.4}
+0: {'loss': 0.8259, 'grad_norm': 0.8122270958513979, 'learning_rate': 2e-05, 'epoch': 0.4}
+0: 725 [2:59:19<4:24:27,  1.78s/it]
                                                        

 40%|███▉      | 5820/14725 [2:59:19<4:24:27,  1.78s/it]
 40%|███▉      | 5821/14725 [2:59:21<4:28:02,  1.81s/it]
 40%|███▉      | 5822/14725 [2:59:23<4:27:54,  1.81s/it]
 40%|███▉      | 5823/14725 [2:59:25<4:28:39,  1.81s/it]
 40%|███▉      | 5824/14725 [2:59:27<4:30:16,  1.82s/it]
 40%|███▉      | 5825/14725 [2:59:28<4:26:45,  1.80s/it]
 40%|███▉      | 5826/14725 [2:59:30<4:28:24,  1.81s/it]
 40%|███▉      | 5827/14725 [2:59:32<4:26:03,  1.79s/it]
 40%|███▉      | 5828/14725 [2:59:34<4:26:53,  1.80s/it]
 40%|███▉      | 5829/14725 [2:59:36<4:27:32,  1.80s/it]
 40%|███▉      | 5830/14725 [2:59:37<4:30:34,  1.83s/it]
                                                        

 40%|███▉      | 5830/14725 [2:59:37<4:30:34,  1.83s/it]
 40%|███▉      | 5831/14725 [2:59:39<4:28:32,  1.81s/it]
 40%|███▉      | 5832/
+0: {'loss': 0.8112, 'grad_norm': 0.8065261005254469, 'learning_rate': 2e-05, 'epoch': 0.4}
+0: 14725 [2:59:41<4:26:20,  1.80s/it]
 40%|███▉      | 5833/14725 [2:59:43<4:26:04,  1.80s/it]
 40%|███▉      | 5834/14725 [2:59:45<4:26:30,  1.80s/it]
 40%|███▉      | 5835/14725 [2:59:46<4:29:12,  1.82s/it]
 40%|███▉      | 5836/14725 [2:59:48<4:27:58,  1.81s/it]
 40%|███▉      | 5837/14725 [2:59:50<4:28:03,  1.81s/it]
 40%|███▉      | 5838/14725 [2:59:52<4:25:35,  1.79s/it]
 40%|███▉      | 5839/14725 [2:59:54<4:25:16,  1.79s/it]
 40%|███▉      | 5840/14725 [2:59:56<4:30:36,  1.83s/it]
                                                        

 40%|███▉      | 5840/14725 [2:59:56<4:30:36,  1.83s/it]
 40%|███▉      | 5841/14725 [2:59:57<4:34:06,  1.85s/it]
 40%|███▉      | 5842/14725 [2:59:59<4:32:16,  1.84s/it]
 40%|███▉      | 5843/14725 [3:00:01<4:28:57,  1.82s/it]
 40%|███▉      | 5844/14725 [3:00:03<4:27:54,  1.81s/it]
 40%|███▉      | 5845/14725 [3:00:05<4:27:47,  1.81s/it]
 40%|███▉    
+0: {'loss': 0.8203, 'grad_norm': 0.7247213914659332, 'learning_rate': 2e-05, 'epoch': 0.4}
+0:   | 5846/14725 [3:00:06<4:28:29,  1.81s/it]
 40%|███▉      | 5847/14725 [3:00:08<4:26:21,  1.80s/it]
 40%|███▉      | 5848/14725 [3:00:10<4:26:42,  1.80s/it]
 40%|███▉      | 5849/14725 [3:00:12<4:25:33,  1.80s/it]
 40%|███▉      | 5850/14725 [3:00:14<4:25:57,  1.80s/it]
                                                        

 40%|███▉      | 5850/14725 [3:00:14<4:25:57,  1.80s/it]
 40%|███▉      | 5851/14725 [3:00:15<4:25:55,  1.80s/it]
 40%|███▉      | 5852/14725 [3:00:17<4:27:49,  1.81s/it]
 40%|███▉      | 5853/14725 [3:00:19<4:25:57,  1.80s/it]
 40%|███▉      | 5854/14725 [3:00:21<4:29:36,  1.82s/it]
 40%|███▉      | 5855/14725 [3:00:23<4:28:29,  1.82s/it]
 40%|███▉      | 5856/14725 [3:00:24<4:27:55,  1.81s/it]
 40%|███▉      | 5857/14725 [3:00:26<4:25:30,  1.80s/it]
 40%|███▉      | 5858/14725 [3:00:28<4:22:55,  1.78s/it]
 40%|███▉      | 5859/14725 [3:00:30<4:25:19,  1.80s/it]
 40%|██�
+0: {'loss': 0.8214, 'grad_norm': 0.7487090684990989, 'learning_rate': 2e-05, 'epoch': 0.4}
+0: {'loss': 0.8373, 'grad_norm': 0.714939004080695, 'learning_rate': 2e-05, 'epoch': 0.4}
+0: ��▉      | 5860/14725 [3:00:32<4:23:18,  1.78s/it]
                                                        

 40%|███▉      | 5860/14725 [3:00:32<4:23:18,  1.78s/it]
 40%|███▉      | 5861/14725 [3:00:33<4:26:28,  1.80s/it]
 40%|███▉      | 5862/14725 [3:00:35<4:30:49,  1.83s/it]
 40%|███▉      | 5863/14725 [3:00:37<4:31:13,  1.84s/it]
 40%|███▉      | 5864/14725 [3:00:39<4:27:14,  1.81s/it]
 40%|███▉      | 5865/14725 [3:00:41<4:25:46,  1.80s/it]
 40%|███▉      | 5866/14725 [3:00:43<4:26:51,  1.81s/it]
 40%|███▉      | 5867/14725 [3:00:44<4:28:51,  1.82s/it]
 40%|███▉      | 5868/14725 [3:00:46<4:25:36,  1.80s/it]
 40%|███▉      | 5869/14725 [3:00:48<4:24:13,  1.79s/it]
 40%|███▉      | 5870/14725 [3:00:50<4:23:05,  1.78s/it]
                                                        

 40%|███▉      | 5870/14725 [3:00:50<4:23:05,  1.78s/it]
 40%|███▉      | 5871/14725 [3:00:51<4:24:24,  1.79s/it]
 40%|█�
+0: {'loss': 0.8261, 'grad_norm': 0.7553259354199151, 'learning_rate': 2e-05, 'epoch': 0.4}
+0: �█▉      | 5872/14725 [3:00:53<4:27:32,  1.81s/it]
 40%|███▉      | 5873/14725 [3:00:55<4:28:10,  1.82s/it]
 40%|███▉      | 5874/14725 [3:00:57<4:27:49,  1.82s/it]
 40%|███▉      | 5875/14725 [3:00:59<4:25:57,  1.80s/it]
 40%|███▉      | 5876/14725 [3:01:01<4:24:57,  1.80s/it]
 40%|███▉      | 5877/14725 [3:01:02<4:26:35,  1.81s/it]
 40%|███▉      | 5878/14725 [3:01:04<4:24:00,  1.79s/it]
 40%|███▉      | 5879/14725 [3:01:06<4:24:51,  1.80s/it]
 40%|███▉      | 5880/14725 [3:01:08<4:24:32,  1.79s/it]
                                                        

 40%|███▉      | 5880/14725 [3:01:08<4:24:32,  1.79s/it]
 40%|███▉      | 5881/14725 [3:01:09<4:24:24,  1.79s/it]
 40%|███▉      | 5882/14725 [3:01:11<4:24:36,  1.80s/it]
 40%|███▉      | 5883/14725 [3:01:13<4:27:49,  1.82s/it]
 40%|███▉      | 5884/14725 [3:01:15<4:24:32,  1.80s/it]
 40%|███▉      | 5885/14725 [3:01:17<4:26:18,  1.81s/it]
 
+0: {'loss': 0.8321, 'grad_norm': 0.841309250378842, 'learning_rate': 2e-05, 'epoch': 0.4}
+0: 40%|███▉      | 5886/14725 [3:01:19<4:26:12,  1.81s/it]
 40%|███▉      | 5887/14725 [3:01:20<4:24:33,  1.80s/it]
 40%|███▉      | 5888/14725 [3:01:22<4:23:30,  1.79s/it]
 40%|███▉      | 5889/14725 [3:01:24<4:27:03,  1.81s/it]
 40%|████      | 5890/14725 [3:01:26<4:25:51,  1.81s/it]
                                                        

 40%|████      | 5890/14725 [3:01:26<4:25:51,  1.81s/it]
 40%|████      | 5891/14725 [3:01:31<6:47:49,  2.77s/it]
 40%|████      | 5892/14725 [3:01:33<6:07:19,  2.50s/it]
 40%|████      | 5893/14725 [3:01:34<5:36:16,  2.28s/it]
 40%|████      | 5894/14725 [3:01:36<5:16:49,  2.15s/it]
 40%|████      | 5895/14725 [3:01:38<4:58:57,  2.03s/it]
 40%|████      | 5896/14725 [3:01:40<4:49:47,  1.97s/it]
 40%|████      | 5897/14725 [3:01:42<4:44:47,  1.94s/it]
 40%|████      | 5898/14725 [3:01:43<4:37:20,  1.89s/it]
 40%|████      | 5899/14725 [3:01:45<4:32:07,  1.
+0: {'loss': 0.8138, 'grad_norm': 0.7969431876084819, 'learning_rate': 2e-05, 'epoch': 0.4}
+0: {'loss': 0.8405, 'grad_norm': 0.7560497809937721, 'learning_rate': 2e-05, 'epoch': 0.4}
+0: 85s/it]
 40%|████      | 5900/14725 [3:01:47<4:29:07,  1.83s/it]
                                                        

 40%|████      | 5900/14725 [3:01:47<4:29:07,  1.83s/it]
 40%|████      | 5901/14725 [3:01:49<4:32:13,  1.85s/it]
 40%|████      | 5902/14725 [3:01:51<4:35:53,  1.88s/it]
 40%|████      | 5903/14725 [3:01:53<4:31:08,  1.84s/it]
 40%|████      | 5904/14725 [3:01:54<4:28:21,  1.83s/it]
 40%|████      | 5905/14725 [3:01:56<4:27:26,  1.82s/it]
 40%|████      | 5906/14725 [3:01:58<4:23:47,  1.79s/it]
 40%|████      | 5907/14725 [3:02:00<4:22:27,  1.79s/it]
 40%|████      | 5908/14725 [3:02:02<4:25:54,  1.81s/it]
 40%|████      | 5909/14725 [3:02:03<4:27:22,  1.82s/it]
 40%|████      | 5910/14725 [3:02:05<4:27:14,  1.82s/it]
                                                        

 40%|████      | 5910/14725 [3:02:05<4:27:14,  1.82s/it]
 40%|████      | 5911/14725 [3:02:07<4:24:54,  
+0: {'loss': 0.8314, 'grad_norm': 0.7451514734507749, 'learning_rate': 2e-05, 'epoch': 0.4}
+0: 1.80s/it]
 40%|████      | 5912/14725 [3:02:09<4:32:57,  1.86s/it]
 40%|████      | 5913/14725 [3:02:11<4:31:51,  1.85s/it]
 40%|████      | 5914/14725 [3:02:13<4:30:29,  1.84s/it]
 40%|████      | 5915/14725 [3:02:15<4:34:16,  1.87s/it]
 40%|████      | 5916/14725 [3:02:16<4:31:46,  1.85s/it]
 40%|████      | 5917/14725 [3:02:18<4:28:46,  1.83s/it]
 40%|████      | 5918/14725 [3:02:20<4:28:24,  1.83s/it]
 40%|████      | 5919/14725 [3:02:22<4:26:33,  1.82s/it]
 40%|████      | 5920/14725 [3:02:24<4:23:39,  1.80s/it]
                                                        

 40%|████      | 5920/14725 [3:02:24<4:23:39,  1.80s/it]
 40%|████      | 5921/14725 [3:02:25<4:22:46,  1.79s/it]
 40%|████      | 5922/14725 [3:02:27<4:21:36,  1.78s/it]
 40%|████      | 5923/14725 [3:02:29<4:21:50,  1.78s/it]
 40%|████      | 5924/14725 [3:02:31<4:23:11,  1.79s/it]
 40%|████      | 5925/14725 [3:02:33<4
+0: {'loss': 0.8068, 'grad_norm': 0.7514039353322204, 'learning_rate': 2e-05, 'epoch': 0.4}
+0: :26:08,  1.81s/it]
 40%|████      | 5926/14725 [3:02:34<4:24:26,  1.80s/it]
 40%|████      | 5927/14725 [3:02:36<4:27:01,  1.82s/it]
 40%|████      | 5928/14725 [3:02:38<4:30:10,  1.84s/it]
 40%|████      | 5929/14725 [3:02:40<4:27:41,  1.83s/it]
 40%|████      | 5930/14725 [3:02:42<4:26:14,  1.82s/it]
                                                        

 40%|████      | 5930/14725 [3:02:42<4:26:14,  1.82s/it]
 40%|████      | 5931/14725 [3:02:43<4:26:27,  1.82s/it]
 40%|████      | 5932/14725 [3:02:45<4:28:02,  1.83s/it]
 40%|████      | 5933/14725 [3:02:47<4:25:55,  1.81s/it]
 40%|████      | 5934/14725 [3:02:49<4:24:41,  1.81s/it]
 40%|████      | 5935/14725 [3:02:51<4:23:22,  1.80s/it]
 40%|████      | 5936/14725 [3:02:52<4:23:44,  1.80s/it]
 40%|████      | 5937/14725 [3:02:54<4:23:03,  1.80s/it]
 40%|████      | 5938/14725 [3:02:56<4:23:50,  1.80s/it]
 40%|████      | 5939/14725 [
+0: {'loss': 0.8354, 'grad_norm': 0.8451637105054961, 'learning_rate': 2e-05, 'epoch': 0.4}
+0: {'loss': 0.8281, 'grad_norm': 0.813675799082084, 'learning_rate': 2e-05, 'epoch': 0.4}
+0: 3:02:58<4:22:30,  1.79s/it]
 40%|████      | 5940/14725 [3:03:00<4:23:32,  1.80s/it]
                                                        

 40%|████      | 5940/14725 [3:03:00<4:23:32,  1.80s/it]
 40%|████      | 5941/14725 [3:03:01<4:22:19,  1.79s/it]
 40%|████      | 5942/14725 [3:03:03<4:20:08,  1.78s/it]
 40%|████      | 5943/14725 [3:03:05<4:23:34,  1.80s/it]
 40%|████      | 5944/14725 [3:03:07<4:24:45,  1.81s/it]
 40%|████      | 5945/14725 [3:03:09<4:28:11,  1.83s/it]
 40%|████      | 5946/14725 [3:03:11<4:25:16,  1.81s/it]
 40%|████      | 5947/14725 [3:03:12<4:29:52,  1.84s/it]
 40%|████      | 5948/14725 [3:03:14<4:30:30,  1.85s/it]
 40%|████      | 5949/14725 [3:03:16<4:26:44,  1.82s/it]
 40%|████      | 5950/14725 [3:03:18<4:25:56,  1.82s/it]
                                                        

 40%|████      | 5950/14725 [3:03:18<4:25:56,  1.82s/it]
 40%|████      | 5951/14725
+0: {'loss': 0.8441, 'grad_norm': 0.67967981614406, 'learning_rate': 2e-05, 'epoch': 0.4}
+0:  [3:03:20<4:25:58,  1.82s/it]
 40%|████      | 5952/14725 [3:03:21<4:22:40,  1.80s/it]
 40%|████      | 5953/14725 [3:03:23<4:22:41,  1.80s/it]
 40%|████      | 5954/14725 [3:03:25<4:21:22,  1.79s/it]
 40%|████      | 5955/14725 [3:03:27<4:22:39,  1.80s/it]
 40%|████      | 5956/14725 [3:03:29<4:21:14,  1.79s/it]
 40%|████      | 5957/14725 [3:03:30<4:22:57,  1.80s/it]
 40%|████      | 5958/14725 [3:03:32<4:20:39,  1.78s/it]
 40%|████      | 5959/14725 [3:03:34<4:19:17,  1.77s/it]
 40%|████      | 5960/14725 [3:03:36<4:18:31,  1.77s/it]
                                                        

 40%|████      | 5960/14725 [3:03:36<4:18:31,  1.77s/it]
 40%|████      | 5961/14725 [3:03:38<4:21:28,  1.79s/it]
 40%|████      | 5962/14725 [3:03:39<4:19:58,  1.78s/it]
 40%|████      | 5963/14725 [3:03:41<4:21:28,  1.79s/it]
 41%|████      | 5964/14725 [3:03:43<4:24:57,  1.81s/it]
 41%|████      | 5
+0: {'loss': 0.8473, 'grad_norm': 0.7027605699328203, 'learning_rate': 2e-05, 'epoch': 0.41}
+0: 965/14725 [3:03:45<4:22:27,  1.80s/it]
 41%|████      | 5966/14725 [3:03:46<4:21:51,  1.79s/it]
 41%|████      | 5967/14725 [3:03:48<4:20:21,  1.78s/it]
 41%|████      | 5968/14725 [3:03:50<4:20:10,  1.78s/it]
 41%|████      | 5969/14725 [3:03:52<4:21:54,  1.79s/it]
 41%|████      | 5970/14725 [3:03:54<4:21:23,  1.79s/it]
                                                        

 41%|████      | 5970/14725 [3:03:54<4:21:23,  1.79s/it]
 41%|████      | 5971/14725 [3:03:55<4:23:19,  1.80s/it]
 41%|████      | 5972/14725 [3:03:57<4:26:45,  1.83s/it]
 41%|████      | 5973/14725 [3:03:59<4:23:21,  1.81s/it]
 41%|████      | 5974/14725 [3:04:01<4:26:32,  1.83s/it]
 41%|████      | 5975/14725 [3:04:03<4:24:31,  1.81s/it]
 41%|████      | 5976/14725 [3:04:05<4:22:42,  1.80s/it]
 41%|████      | 5977/14725 [3:04:06<4:22:45,  1.80s/it]
 41%|████      | 5978/14725 [3:04:08<4:22:05,  1.80s/it]
 41%|████
+0: {'loss': 0.851, 'grad_norm': 0.7162283445031186, 'learning_rate': 2e-05, 'epoch': 0.41}
+0: {'loss': 0.8259, 'grad_norm': 0.7310276329361557, 'learning_rate': 2e-05, 'epoch': 0.41}
+0:       | 5979/14725 [3:04:10<4:20:02,  1.78s/it]
 41%|████      | 5980/14725 [3:04:12<4:18:53,  1.78s/it]
                                                        

 41%|████      | 5980/14725 [3:04:12<4:18:53,  1.78s/it]
 41%|████      | 5981/14725 [3:04:14<4:24:38,  1.82s/it]
 41%|████      | 5982/14725 [3:04:15<4:23:13,  1.81s/it]
 41%|████      | 5983/14725 [3:04:17<4:22:31,  1.80s/it]
 41%|████      | 5984/14725 [3:04:19<4:23:05,  1.81s/it]
 41%|████      | 5985/14725 [3:04:21<4:21:52,  1.80s/it]
 41%|████      | 5986/14725 [3:04:23<4:26:17,  1.83s/it]
 41%|████      | 5987/14725 [3:04:25<4:28:53,  1.85s/it]
 41%|████      | 5988/14725 [3:04:26<4:28:54,  1.85s/it]
 41%|████      | 5989/14725 [3:04:28<4:29:45,  1.85s/it]
 41%|████      | 5990/14725 [3:04:30<4:26:52,  1.83s/it]
                                                        

 41%|████      | 5990/14725 [3:04:30<4:26:52,  1.83s/it]
 41%|███�
+0: {'loss': 0.8336, 'grad_norm': 0.72431625315181, 'learning_rate': 2e-05, 'epoch': 0.41}
+0: ��      | 5991/14725 [3:04:32<4:25:59,  1.83s/it]
 41%|████      | 5992/14725 [3:04:34<4:24:02,  1.81s/it]
 41%|████      | 5993/14725 [3:04:35<4:24:21,  1.82s/it]
 41%|████      | 5994/14725 [3:04:37<4:26:00,  1.83s/it]
 41%|████      | 5995/14725 [3:04:39<4:37:58,  1.91s/it]
 41%|████      | 5996/14725 [3:04:41<4:34:55,  1.89s/it]
 41%|████      | 5997/14725 [3:04:43<4:31:28,  1.87s/it]
 41%|████      | 5998/14725 [3:04:45<4:28:46,  1.85s/it]
 41%|████      | 5999/14725 [3:04:47<4:24:59,  1.82s/it]
 41%|████      | 6000/14725 [3:04:48<4:22:48,  1.81s/it]
                                                        

 41%|████      | 6000/14725 [3:04:48<4:22:48,  1.81s/it]
 41%|████      | 6001/14725 [3:04:50<4:25:40,  1.83s/it]
 41%|████      | 6002/14725 [3:04:52<4:24:44,  1.82s/it]
 41%|████      | 6003/14725 [3:04:54<4:25:25,  1.83s/it]
 41%|████      | 6004/14725 [3:04:56<4:24:34,  1.82s/it]
 41%|�
+0: {'loss': 0.8335, 'grad_norm': 0.7181568349443875, 'learning_rate': 2e-05, 'epoch': 0.41}
+0: ��███      | 6005/14725 [3:04:58<4:26:07,  1.83s/it]
 41%|████      | 6006/14725 [3:04:59<4:23:48,  1.82s/it]
 41%|████      | 6007/14725 [3:05:01<4:27:11,  1.84s/it]
 41%|████      | 6008/14725 [3:05:03<4:27:25,  1.84s/it]
 41%|████      | 6009/14725 [3:05:05<4:26:24,  1.83s/it]
 41%|████      | 6010/14725 [3:05:07<4:27:10,  1.84s/it]
                                                        

 41%|████      | 6010/14725 [3:05:07<4:27:10,  1.84s/it]
 41%|████      | 6011/14725 [3:05:09<4:25:51,  1.83s/it]
 41%|████      | 6012/14725 [3:05:10<4:24:40,  1.82s/it]
 41%|████      | 6013/14725 [3:05:12<4:22:31,  1.81s/it]
 41%|████      | 6014/14725 [3:05:14<4:23:01,  1.81s/it]
 41%|████      | 6015/14725 [3:05:16<4:23:30,  1.82s/it]
 41%|████      | 6016/14725 [3:05:18<4:22:18,  1.81s/it]
 41%|████      | 6017/14725 [3:05:19<4:22:52,  1.81s/it]
 41%|████      | 6018/14725 [3:05:21<4:28:29,  1.85s/i
+0: {'loss': 0.8321, 'grad_norm': 0.8107592022009851, 'learning_rate': 2e-05, 'epoch': 0.41}
+0: {'loss': 0.8307, 'grad_norm': 0.7216529473334784, 'learning_rate': 2e-05, 'epoch': 0.41}
+0: t]
 41%|████      | 6019/14725 [3:05:23<4:24:28,  1.82s/it]
 41%|████      | 6020/14725 [3:05:25<4:24:17,  1.82s/it]
                                                        

 41%|████      | 6020/14725 [3:05:25<4:24:17,  1.82s/it]
 41%|████      | 6021/14725 [3:05:27<4:22:38,  1.81s/it]
 41%|████      | 6022/14725 [3:05:29<4:24:32,  1.82s/it]
 41%|████      | 6023/14725 [3:05:30<4:26:18,  1.84s/it]
 41%|████      | 6024/14725 [3:05:32<4:25:17,  1.83s/it]
 41%|████      | 6025/14725 [3:05:34<4:22:06,  1.81s/it]
 41%|████      | 6026/14725 [3:05:36<4:20:02,  1.79s/it]
 41%|████      | 6027/14725 [3:05:38<4:21:50,  1.81s/it]
 41%|████      | 6028/14725 [3:05:39<4:22:27,  1.81s/it]
 41%|████      | 6029/14725 [3:05:42<4:44:58,  1.97s/it]
 41%|████      | 6030/14725 [3:05:44<4:36:32,  1.91s/it]
                                                        

 41%|████      | 6030/14725 [3:05:44<4:36:32,  1.91s
+0: {'loss': 0.8357, 'grad_norm': 0.8237413110360493, 'learning_rate': 2e-05, 'epoch': 0.41}
+0: /it]
 41%|████      | 6031/14725 [3:05:45<4:32:00,  1.88s/it]
 41%|████      | 6032/14725 [3:05:47<4:30:43,  1.87s/it]
 41%|████      | 6033/14725 [3:05:49<4:30:51,  1.87s/it]
 41%|████      | 6034/14725 [3:05:51<4:25:03,  1.83s/it]
 41%|████      | 6035/14725 [3:05:53<4:23:40,  1.82s/it]
 41%|████      | 6036/14725 [3:05:54<4:21:53,  1.81s/it]
 41%|████      | 6037/14725 [3:05:56<4:22:21,  1.81s/it]
 41%|████      | 6038/14725 [3:05:58<4:23:01,  1.82s/it]
 41%|████      | 6039/14725 [3:06:00<4:22:05,  1.81s/it]
 41%|████      | 6040/14725 [3:06:02<4:22:12,  1.81s/it]
                                                        

 41%|████      | 6040/14725 [3:06:02<4:22:12,  1.81s/it]
 41%|████      | 6041/14725 [3:06:03<4:21:32,  1.81s/it]
 41%|████      | 6042/14725 [3:06:05<4:18:51,  1.79s/it]
 41%|████      | 6043/14725 [3:06:07<4:19:22,  1.79s/it]
 41%|████      | 6044/14725 [3:06:09<4:19:4
+0: {'loss': 0.823, 'grad_norm': 0.73667338040018, 'learning_rate': 2e-05, 'epoch': 0.41}
+0: 5,  1.80s/it]
 41%|████      | 6045/14725 [3:06:11<4:20:16,  1.80s/it]
 41%|████      | 6046/14725 [3:06:12<4:18:47,  1.79s/it]
 41%|████      | 6047/14725 [3:06:14<4:18:51,  1.79s/it]
 41%|████      | 6048/14725 [3:06:16<4:35:32,  1.91s/it]
 41%|████      | 6049/14725 [3:06:18<4:31:06,  1.87s/it]
 41%|████      | 6050/14725 [3:06:20<4:25:42,  1.84s/it]
                                                        

 41%|████      | 6050/14725 [3:06:20<4:25:42,  1.84s/it]
 41%|████      | 6051/14725 [3:06:22<4:25:58,  1.84s/it]
 41%|████      | 6052/14725 [3:06:24<4:39:38,  1.93s/it]
 41%|████      | 6053/14725 [3:06:26<4:33:39,  1.89s/it]
 41%|████      | 6054/14725 [3:06:27<4:28:55,  1.86s/it]
 41%|████      | 6055/14725 [3:06:29<4:28:40,  1.86s/it]
 41%|████      | 6056/14725 [3:06:31<4:25:36,  1.84s/it]
 41%|████      | 6057/14725 [3:06:33<4:24:08,  1.83s/it]
 41%|████      | 6058/14725 [3:06:
+0: {'loss': 0.8393, 'grad_norm': 0.7482451357693088, 'learning_rate': 2e-05, 'epoch': 0.41}
+0: {'loss': 0.8266, 'grad_norm': 0.7303342797220427, 'learning_rate': 2e-05, 'epoch': 0.41}
+0: 35<4:21:15,  1.81s/it]
 41%|████      | 6059/14725 [3:06:36<4:22:56,  1.82s/it]
 41%|████      | 6060/14725 [3:06:38<4:26:32,  1.85s/it]
                                                        

 41%|████      | 6060/14725 [3:06:38<4:26:32,  1.85s/it]
 41%|████      | 6061/14725 [3:06:40<4:29:46,  1.87s/it]
 41%|████      | 6062/14725 [3:06:42<4:25:31,  1.84s/it]
 41%|████      | 6063/14725 [3:06:44<4:26:18,  1.84s/it]
 41%|████      | 6064/14725 [3:06:46<4:22:32,  1.82s/it]
 41%|████      | 6065/14725 [3:06:48<4:22:06,  1.82s/it]
 41%|████      | 6066/14725 [3:06:49<4:20:31,  1.81s/it]
 41%|████      | 6067/14725 [3:06:51<4:19:26,  1.80s/it]
 41%|████      | 6068/14725 [3:06:53<4:17:01,  1.78s/it]
 41%|████      | 6069/14725 [3:06:55<4:19:57,  1.80s/it]
 41%|████      | 6070/14725 [3:06:56<4:20:56,  1.81s/it]
                                                        

 41%|████      | 6070/14725 [3:0
+0: {'loss': 0.8391, 'grad_norm': 0.8105634437123138, 'learning_rate': 2e-05, 'epoch': 0.41}
+0: 6:56<4:20:56,  1.81s/it]
 41%|████      | 6071/14725 [3:06:58<4:25:30,  1.84s/it]
 41%|████      | 6072/14725 [3:07:00<4:24:46,  1.84s/it]
 41%|████      | 6073/14725 [3:07:02<4:23:50,  1.83s/it]
 41%|████      | 6074/14725 [3:07:04<4:22:16,  1.82s/it]
 41%|████▏     | 6075/14725 [3:07:06<4:22:15,  1.82s/it]
 41%|████▏     | 6076/14725 [3:07:07<4:23:17,  1.83s/it]
 41%|████▏     | 6077/14725 [3:07:10<4:39:12,  1.94s/it]
 41%|████▏     | 6078/14725 [3:07:12<4:34:28,  1.90s/it]
 41%|████▏     | 6079/14725 [3:07:13<4:31:04,  1.88s/it]
 41%|████▏     | 6080/14725 [3:07:15<4:30:07,  1.87s/it]
                                                        

 41%|████▏     | 6080/14725 [3:07:15<4:30:07,  1.87s/it]
 41%|████▏     | 6081/14725 [3:07:17<4:26:15,  1.85s/it]
 41%|████▏     | 6082/14725 [3:07:19<4:24:23,  1.84s/it]
 41%|████▏     | 6083/14725 [3:07:21<4:20:29,  1.81s/it]
 41%|██
+0: {'loss': 0.8298, 'grad_norm': 0.7167066607829053, 'learning_rate': 2e-05, 'epoch': 0.41}
+0: ██▏     | 6084/14725 [3:07:22<4:22:25,  1.82s/it]
 41%|████▏     | 6085/14725 [3:07:24<4:22:39,  1.82s/it]
 41%|████▏     | 6086/14725 [3:07:26<4:22:51,  1.83s/it]
 41%|████▏     | 6087/14725 [3:07:28<4:20:13,  1.81s/it]
 41%|████▏     | 6088/14725 [3:07:30<4:20:01,  1.81s/it]
 41%|████▏     | 6089/14725 [3:07:31<4:18:52,  1.80s/it]
 41%|████▏     | 6090/14725 [3:07:33<4:19:18,  1.80s/it]
                                                        

 41%|████▏     | 6090/14725 [3:07:33<4:19:18,  1.80s/it]
 41%|████▏     | 6091/14725 [3:07:35<4:17:26,  1.79s/it]
 41%|████▏     | 6092/14725 [3:07:37<4:17:02,  1.79s/it]
 41%|████▏     | 6093/14725 [3:07:39<4:17:01,  1.79s/it]
 41%|████▏     | 6094/14725 [3:07:40<4:19:02,  1.80s/it]
 41%|████▏     | 6095/14725 [3:07:42<4:18:06,  1.79s/it]
 41%|████▏     | 6096/14725 [3:07:44<4:20:06,  1.81s/it]
 41%|████▏     | 6097/14725 [
+0: {'loss': 0.8357, 'grad_norm': 0.7785180763654852, 'learning_rate': 2e-05, 'epoch': 0.41}
+0: 3:07:46<4:21:33,  1.82s/it]
 41%|████▏     | 6098/14725 [3:07:48<4:23:29,  1.83s/it]
 41%|████▏     | 6099/14725 [3:07:50<4:22:41,  1.83s/it]
 41%|████▏     | 6100/14725 [3:07:51<4:21:26,  1.82s/it]
                                                        

 41%|████▏     | 6100/14725 [3:07:51<4:21:26,  1.82s/it]
 41%|████▏     | 6101/14725 [3:07:53<4:21:05,  1.82s/it]
 41%|████▏     | 6102/14725 [3:07:55<4:19:38,  1.81s/it]
 41%|████▏     | 6103/14725 [3:07:57<4:18:23,  1.80s/it]
 41%|████▏     | 6104/14725 [3:07:58<4:18:21,  1.80s/it]
 41%|████▏     | 6105/14725 [3:08:00<4:18:34,  1.80s/it]
 41%|████▏     | 6106/14725 [3:08:02<4:16:33,  1.79s/it]
 41%|████▏     | 6107/14725 [3:08:04<4:15:14,  1.78s/it]
 41%|████▏     | 6108/14725 [3:08:06<4:14:31,  1.77s/it]
 41%|████▏     | 6109/14725 [3:08:07<4:17:16,  1.79s/it]
 41%|████▏     | 6110/14725 [3:08:09<4:17:22,  1.79s/it]
+0: {'loss': 0.8278, 'grad_norm': 0.8092565233168497, 'learning_rate': 2e-05, 'epoch': 0.41}
+0: {'loss': 0.8572, 'grad_norm': 0.7480480858578136, 'learning_rate': 2e-05, 'epoch': 0.42}
+0:                                                         

 41%|████▏     | 6110/14725 [3:08:09<4:17:22,  1.79s/it]
 42%|████▏     | 6111/14725 [3:08:11<4:21:49,  1.82s/it]
 42%|████▏     | 6112/14725 [3:08:13<4:22:44,  1.83s/it]
 42%|████▏     | 6113/14725 [3:08:15<4:20:11,  1.81s/it]
 42%|████▏     | 6114/14725 [3:08:16<4:17:53,  1.80s/it]
 42%|████▏     | 6115/14725 [3:08:18<4:18:57,  1.80s/it]
 42%|████▏     | 6116/14725 [3:08:20<4:16:43,  1.79s/it]
 42%|████▏     | 6117/14725 [3:08:22<4:16:59,  1.79s/it]
 42%|████▏     | 6118/14725 [3:08:24<4:20:56,  1.82s/it]
 42%|████▏     | 6119/14725 [3:08:25<4:18:08,  1.80s/it]
 42%|████▏     | 6120/14725 [3:08:27<4:16:01,  1.79s/it]
                                                        

 42%|████▏     | 6120/14725 [3:08:27<4:16:01,  1.79s/it]
 42%|████▏     | 6121/14725 [3:08:29<4:15:30,  1.78s/it]
 42%|████▏     | 6122/14725
+0: {'loss': 0.8241, 'grad_norm': 0.694420484589128, 'learning_rate': 2e-05, 'epoch': 0.42}
+0:  [3:08:31<4:23:51,  1.84s/it]
 42%|████▏     | 6123/14725 [3:08:33<4:25:04,  1.85s/it]
 42%|████▏     | 6124/14725 [3:08:35<4:21:32,  1.82s/it]
 42%|████▏     | 6125/14725 [3:08:36<4:21:29,  1.82s/it]
 42%|████▏     | 6126/14725 [3:08:38<4:19:30,  1.81s/it]
 42%|████▏     | 6127/14725 [3:08:40<4:17:57,  1.80s/it]
 42%|████▏     | 6128/14725 [3:08:42<4:16:53,  1.79s/it]
 42%|████▏     | 6129/14725 [3:08:44<4:19:51,  1.81s/it]
 42%|████▏     | 6130/14725 [3:08:45<4:18:06,  1.80s/it]
                                                        

 42%|████▏     | 6130/14725 [3:08:45<4:18:06,  1.80s/it]
 42%|████▏     | 6131/14725 [3:08:47<4:17:11,  1.80s/it]
 42%|████▏     | 6132/14725 [3:08:49<4:17:40,  1.80s/it]
 42%|████▏     | 6133/14725 [3:08:51<4:16:49,  1.79s/it]
 42%|████▏     | 6134/14725 [3:08:53<4:19:12,  1.81s/it]
 42%|████▏     | 6135/14725 [3:08:54<4:18:03,  1.80s/it
+0: {'loss': 0.8209, 'grad_norm': 0.7160188602144699, 'learning_rate': 2e-05, 'epoch': 0.42}
+0: ]
 42%|████▏     | 6136/14725 [3:08:56<4:16:07,  1.79s/it]
 42%|████▏     | 6137/14725 [3:08:58<4:16:33,  1.79s/it]
 42%|████▏     | 6138/14725 [3:09:00<4:18:29,  1.81s/it]
 42%|████▏     | 6139/14725 [3:09:02<4:19:42,  1.81s/it]
 42%|████▏     | 6140/14725 [3:09:03<4:20:47,  1.82s/it]
                                                        

 42%|████▏     | 6140/14725 [3:09:03<4:20:47,  1.82s/it]
 42%|████▏     | 6141/14725 [3:09:05<4:20:54,  1.82s/it]
 42%|████▏     | 6142/14725 [3:09:07<4:22:20,  1.83s/it]
 42%|████▏     | 6143/14725 [3:09:09<4:32:46,  1.91s/it]
 42%|████▏     | 6144/14725 [3:09:11<4:32:11,  1.90s/it]
 42%|████▏     | 6145/14725 [3:09:13<4:28:43,  1.88s/it]
 42%|████▏     | 6146/14725 [3:09:15<4:25:23,  1.86s/it]
 42%|████▏     | 6147/14725 [3:09:17<4:23:43,  1.84s/it]
 42%|████▏     | 6148/14725 [3:09:18<4:22:34,  1.84s/it]
 42%|████▏     |
+0: {'loss': 0.8439, 'grad_norm': 0.7395846280230036, 'learning_rate': 2e-05, 'epoch': 0.42}
+0: {'loss': 0.8326, 'grad_norm': 0.7668126176584721, 'learning_rate': 2e-05, 'epoch': 0.42}
+0:  6149/14725 [3:09:20<4:19:45,  1.82s/it]
 42%|████▏     | 6150/14725 [3:09:22<4:16:46,  1.80s/it]
                                                        

 42%|████▏     | 6150/14725 [3:09:22<4:16:46,  1.80s/it]
 42%|████▏     | 6151/14725 [3:09:24<4:16:00,  1.79s/it]
 42%|████▏     | 6152/14725 [3:09:26<4:16:32,  1.80s/it]
 42%|████▏     | 6153/14725 [3:09:27<4:14:51,  1.78s/it]
 42%|████▏     | 6154/14725 [3:09:29<4:15:37,  1.79s/it]
 42%|████▏     | 6155/14725 [3:09:31<4:20:06,  1.82s/it]
 42%|████▏     | 6156/14725 [3:09:33<4:17:00,  1.80s/it]
 42%|████▏     | 6157/14725 [3:09:35<4:18:38,  1.81s/it]
 42%|████▏     | 6158/14725 [3:09:36<4:20:08,  1.82s/it]
 42%|████▏     | 6159/14725 [3:09:38<4:18:26,  1.81s/it]
 42%|████▏     | 6160/14725 [3:09:40<4:20:34,  1.83s/it]
                                                        

 42%|████▏     | 6160/14725 [3:09:40<4:20:34,  1.83s/
+0: {'loss': 0.8207, 'grad_norm': 0.7677776845070038, 'learning_rate': 2e-05, 'epoch': 0.42}
+0: it]
 42%|████▏     | 6161/14725 [3:09:42<4:18:59,  1.81s/it]
 42%|████▏     | 6162/14725 [3:09:44<4:18:03,  1.81s/it]
 42%|████▏     | 6163/14725 [3:09:45<4:17:00,  1.80s/it]
 42%|████▏     | 6164/14725 [3:09:47<4:15:14,  1.79s/it]
 42%|████▏     | 6165/14725 [3:09:49<4:13:59,  1.78s/it]
 42%|████▏     | 6166/14725 [3:09:51<4:16:34,  1.80s/it]
 42%|████▏     | 6167/14725 [3:09:53<4:15:19,  1.79s/it]
 42%|████▏     | 6168/14725 [3:09:54<4:16:14,  1.80s/it]
 42%|████▏     | 6169/14725 [3:09:56<4:14:44,  1.79s/it]
 42%|████▏     | 6170/14725 [3:09:58<4:17:05,  1.80s/it]
                                                        

 42%|████▏     | 6170/14725 [3:09:58<4:17:05,  1.80s/it]
 42%|████▏     | 6171/14725 [3:10:00<4:16:52,  1.80s/it]
 42%|████▏     | 6172/14725 [3:10:02<4:14:30,  1.79s/it]
 42%|████▏     | 6173/14725 [3:10:03<4:15:48,  1.79s/it]
 42%|████▏    
+0: {'loss': 0.8205, 'grad_norm': 0.7933920201792548, 'learning_rate': 2e-05, 'epoch': 0.42}
+0:  | 6174/14725 [3:10:05<4:17:50,  1.81s/it]
 42%|████▏     | 6175/14725 [3:10:07<4:17:14,  1.81s/it]
 42%|████▏     | 6176/14725 [3:10:09<4:18:55,  1.82s/it]
 42%|████▏     | 6177/14725 [3:10:11<4:21:59,  1.84s/it]
 42%|████▏     | 6178/14725 [3:10:12<4:18:03,  1.81s/it]
 42%|████▏     | 6179/14725 [3:10:14<4:17:45,  1.81s/it]
 42%|████▏     | 6180/14725 [3:10:16<4:17:55,  1.81s/it]
                                                        

 42%|████▏     | 6180/14725 [3:10:16<4:17:55,  1.81s/it]
 42%|████▏     | 6181/14725 [3:10:18<4:17:59,  1.81s/it]
 42%|████▏     | 6182/14725 [3:10:20<4:18:43,  1.82s/it]
 42%|████▏     | 6183/14725 [3:10:21<4:16:52,  1.80s/it]
 42%|████▏     | 6184/14725 [3:10:23<4:17:23,  1.81s/it]
 42%|████▏     | 6185/14725 [3:10:25<4:16:29,  1.80s/it]
 42%|████▏     | 6186/14725 [3:10:27<4:20:51,  1.83s/it]
 42%|████▏     | 6187/14725 [3:10:29<4:20:
+0: {'loss': 0.8292, 'grad_norm': 0.8228022933722688, 'learning_rate': 2e-05, 'epoch': 0.42}
+0: 38,  1.83s/it]
 42%|████▏     | 6188/14725 [3:10:31<4:19:29,  1.82s/it]
 42%|████▏     | 6189/14725 [3:10:32<4:18:17,  1.82s/it]
 42%|████▏     | 6190/14725 [3:10:34<4:15:10,  1.79s/it]
                                                        

 42%|████▏     | 6190/14725 [3:10:34<4:15:10,  1.79s/it]
 42%|████▏     | 6191/14725 [3:10:36<4:13:23,  1.78s/it]
 42%|████▏     | 6192/14725 [3:10:38<4:14:57,  1.79s/it]
 42%|████▏     | 6193/14725 [3:10:39<4:13:34,  1.78s/it]
 42%|████▏     | 6194/14725 [3:10:41<4:14:29,  1.79s/it]
 42%|████▏     | 6195/14725 [3:10:43<4:13:50,  1.79s/it]
 42%|████▏     | 6196/14725 [3:10:45<4:16:05,  1.80s/it]
 42%|████▏     | 6197/14725 [3:10:47<4:15:19,  1.80s/it]
 42%|████▏     | 6198/14725 [3:10:48<4:14:12,  1.79s/it]
 42%|████▏     | 6199/14725 [3:10:50<4:13:20,  1.78s/it]
 42%|████▏     | 6200/14725 [3:10:52<4:12:51,  1.78s/it]
             
+0: {'loss': 0.8339, 'grad_norm': 0.7455053169876522, 'learning_rate': 2e-05, 'epoch': 0.42}
+0: {'loss': 0.8311, 'grad_norm': 0.7024117938254696, 'learning_rate': 2e-05, 'epoch': 0.42}
+0:                                            

 42%|████▏     | 6200/14725 [3:10:52<4:12:51,  1.78s/it]
 42%|████▏     | 6201/14725 [3:10:54<4:13:51,  1.79s/it]
 42%|████▏     | 6202/14725 [3:10:56<4:14:10,  1.79s/it]
 42%|████▏     | 6203/14725 [3:10:57<4:16:27,  1.81s/it]
 42%|████▏     | 6204/14725 [3:10:59<4:15:26,  1.80s/it]
 42%|████▏     | 6205/14725 [3:11:01<4:13:57,  1.79s/it]
 42%|████▏     | 6206/14725 [3:11:03<4:13:13,  1.78s/it]
 42%|████▏     | 6207/14725 [3:11:05<4:11:34,  1.77s/it]
 42%|████▏     | 6208/14725 [3:11:06<4:11:45,  1.77s/it]
 42%|████▏     | 6209/14725 [3:11:08<4:13:49,  1.79s/it]
 42%|████▏     | 6210/14725 [3:11:10<4:15:11,  1.80s/it]
                                                        

 42%|████▏     | 6210/14725 [3:11:10<4:15:11,  1.80s/it]
 42%|████▏     | 6211/14725 [3:11:12<4:18:57,  1.82s/it]
 42%|████▏     | 6212/14725 [3:11:14<4:1
+0: {'loss': 0.8251, 'grad_norm': 0.6927358355995487, 'learning_rate': 2e-05, 'epoch': 0.42}
+0: 8:40,  1.82s/it]
 42%|████▏     | 6213/14725 [3:11:15<4:19:16,  1.83s/it]
 42%|████▏     | 6214/14725 [3:11:17<4:20:53,  1.84s/it]
 42%|████▏     | 6215/14725 [3:11:19<4:19:32,  1.83s/it]
 42%|████▏     | 6216/14725 [3:11:21<4:21:13,  1.84s/it]
 42%|████▏     | 6217/14725 [3:11:23<4:19:06,  1.83s/it]
 42%|████▏     | 6218/14725 [3:11:25<4:18:29,  1.82s/it]
 42%|████▏     | 6219/14725 [3:11:26<4:16:20,  1.81s/it]
 42%|████▏     | 6220/14725 [3:11:28<4:17:01,  1.81s/it]
                                                        

 42%|████▏     | 6220/14725 [3:11:28<4:17:01,  1.81s/it]
 42%|████▏     | 6221/14725 [3:11:30<4:22:06,  1.85s/it]
 42%|████▏     | 6222/14725 [3:11:32<4:18:56,  1.83s/it]
 42%|████▏     | 6223/14725 [3:11:34<4:17:47,  1.82s/it]
 42%|████▏     | 6224/14725 [3:11:35<4:14:55,  1.80s/it]
 42%|████▏     | 6225/14725 [3:11:37<4:14:15,  1.79s/it]
 42%|██
+0: {'loss': 0.8418, 'grad_norm': 0.7876955830896124, 'learning_rate': 2e-05, 'epoch': 0.42}
+0: ██▏     | 6226/14725 [3:11:39<4:13:12,  1.79s/it]
 42%|████▏     | 6227/14725 [3:11:41<4:14:45,  1.80s/it]
 42%|████▏     | 6228/14725 [3:11:43<4:14:24,  1.80s/it]
 42%|████▏     | 6229/14725 [3:11:45<4:17:55,  1.82s/it]
 42%|████▏     | 6230/14725 [3:11:46<4:19:54,  1.84s/it]
                                                        

 42%|████▏     | 6230/14725 [3:11:46<4:19:54,  1.84s/it]
 42%|████▏     | 6231/14725 [3:11:48<4:19:35,  1.83s/it]
 42%|████▏     | 6232/14725 [3:11:50<4:16:54,  1.81s/it]
 42%|████▏     | 6233/14725 [3:11:52<4:15:26,  1.80s/it]
 42%|████▏     | 6234/14725 [3:11:54<4:12:48,  1.79s/it]
 42%|████▏     | 6235/14725 [3:11:55<4:13:08,  1.79s/it]
 42%|████▏     | 6236/14725 [3:11:57<4:11:50,  1.78s/it]
 42%|████▏     | 6237/14725 [3:11:59<4:10:58,  1.77s/it]
 42%|████▏     | 6238/14725 [3:12:01<4:11:00,  1.77s/it]
 42%|████▏     | 6239/14725 [
+0: {'loss': 0.8057, 'grad_norm': 0.7209304224451117, 'learning_rate': 2e-05, 'epoch': 0.42}
+0: {'loss': 0.8464, 'grad_norm': 0.7480391261218067, 'learning_rate': 2e-05, 'epoch': 0.42}
+0: 3:12:02<4:09:21,  1.76s/it]
 42%|████▏     | 6240/14725 [3:12:04<4:09:22,  1.76s/it]
                                                        

 42%|████▏     | 6240/14725 [3:12:04<4:09:22,  1.76s/it]
 42%|████▏     | 6241/14725 [3:12:06<4:12:15,  1.78s/it]
 42%|████▏     | 6242/14725 [3:12:08<4:20:52,  1.85s/it]
 42%|████▏     | 6243/14725 [3:12:10<4:20:03,  1.84s/it]
 42%|████▏     | 6244/14725 [3:12:12<4:18:51,  1.83s/it]
 42%|████▏     | 6245/14725 [3:12:13<4:18:13,  1.83s/it]
 42%|████▏     | 6246/14725 [3:12:15<4:18:40,  1.83s/it]
 42%|████▏     | 6247/14725 [3:12:17<4:19:28,  1.84s/it]
 42%|████▏     | 6248/14725 [3:12:19<4:21:25,  1.85s/it]
 42%|████▏     | 6249/14725 [3:12:21<4:18:47,  1.83s/it]
 42%|████▏     | 6250/14725 [3:12:23<4:18:02,  1.83s/it]
                                                        

 42%|████▏     | 6250/14725 [3:12:23<4:18:02,  1.83s/it]
 42%|█�
+0: {'loss': 0.8169, 'grad_norm': 0.7193742047255243, 'learning_rate': 2e-05, 'epoch': 0.43}
+0: ��██▏     | 6251/14725 [3:12:24<4:17:07,  1.82s/it]
 42%|████▏     | 6252/14725 [3:12:26<4:16:22,  1.82s/it]
 42%|████▏     | 6253/14725 [3:12:28<4:19:07,  1.84s/it]
 42%|████▏     | 6254/14725 [3:12:30<4:19:13,  1.84s/it]
 42%|████▏     | 6255/14725 [3:12:32<4:22:07,  1.86s/it]
 42%|████▏     | 6256/14725 [3:12:34<4:20:14,  1.84s/it]
 42%|████▏     | 6257/14725 [3:12:35<4:18:14,  1.83s/it]
 42%|████▏     | 6258/14725 [3:12:37<4:17:04,  1.82s/it]
 43%|████▎     | 6259/14725 [3:12:39<4:14:04,  1.80s/it]
 43%|████▎     | 6260/14725 [3:12:41<4:12:13,  1.79s/it]
                                                        

 43%|████▎     | 6260/14725 [3:12:41<4:12:13,  1.79s/it]
 43%|████▎     | 6261/14725 [3:12:43<4:13:06,  1.79s/it]
 43%|████▎     | 6262/14725 [3:12:44<4:11:11,  1.78s/it]
 43%|████▎     | 6263/14725 [3:12:46<4:12:26,  1.79s/it]
 43%|████▎     | 6264/14725
+0: {'loss': 0.8229, 'grad_norm': 0.7074684736137485, 'learning_rate': 2e-05, 'epoch': 0.43}
+0:  [3:12:48<4:10:55,  1.78s/it]
 43%|████▎     | 6265/14725 [3:12:50<4:14:41,  1.81s/it]
 43%|████▎     | 6266/14725 [3:12:52<4:17:39,  1.83s/it]
 43%|████▎     | 6267/14725 [3:12:53<4:16:53,  1.82s/it]
 43%|████▎     | 6268/14725 [3:12:55<4:15:46,  1.81s/it]
 43%|████▎     | 6269/14725 [3:12:57<4:14:18,  1.80s/it]
 43%|████▎     | 6270/14725 [3:12:59<4:13:35,  1.80s/it]
                                                        

 43%|████▎     | 6270/14725 [3:12:59<4:13:35,  1.80s/it]
 43%|████▎     | 6271/14725 [3:13:01<4:14:47,  1.81s/it]
 43%|████▎     | 6272/14725 [3:13:02<4:13:56,  1.80s/it]
 43%|████▎     | 6273/14725 [3:13:04<4:11:02,  1.78s/it]
 43%|████▎     | 6274/14725 [3:13:06<4:14:02,  1.80s/it]
 43%|████▎     | 6275/14725 [3:13:08<4:17:56,  1.83s/it]
 43%|████▎     | 6276/14725 [3:13:10<4:16:01,  1.82s/it]
 43%|████▎     | 6277/14725 [3:13:11<4:14:59,  1.81s/it
+0: {'loss': 0.8306, 'grad_norm': 0.6913776076366063, 'learning_rate': 2e-05, 'epoch': 0.43}
+0: ]
 43%|████▎     | 6278/14725 [3:13:13<4:12:33,  1.79s/it]
 43%|████▎     | 6279/14725 [3:13:15<4:10:19,  1.78s/it]
 43%|████▎     | 6280/14725 [3:13:17<4:11:42,  1.79s/it]
                                                        

 43%|████▎     | 6280/14725 [3:13:17<4:11:42,  1.79s/it]
 43%|████▎     | 6281/14725 [3:13:19<4:12:48,  1.80s/it]
 43%|████▎     | 6282/14725 [3:13:20<4:12:58,  1.80s/it]
 43%|████▎     | 6283/14725 [3:13:22<4:14:58,  1.81s/it]
 43%|████▎     | 6284/14725 [3:13:24<4:11:45,  1.79s/it]
 43%|████▎     | 6285/14725 [3:13:26<4:11:30,  1.79s/it]
 43%|████▎     | 6286/14725 [3:13:28<4:14:00,  1.81s/it]
 43%|████▎     | 6287/14725 [3:13:29<4:10:39,  1.78s/it]
 43%|████▎     | 6288/14725 [3:13:31<4:12:37,  1.80s/it]
 43%|████▎     | 6289/14725 [3:13:33<4:11:34,  1.79s/it]
 43%|████▎     | 6290/14725 [3:13:35<4:17:18,  1.83s/it]
                          
+0: {'loss': 0.8295, 'grad_norm': 0.7888280725197427, 'learning_rate': 2e-05, 'epoch': 0.43}
+0: {'loss': 0.8278, 'grad_norm': 0.7420356533382174, 'learning_rate': 2e-05, 'epoch': 0.43}
+0:                               

 43%|████▎     | 6290/14725 [3:13:35<4:17:18,  1.83s/it]
 43%|████▎     | 6291/14725 [3:13:37<4:16:09,  1.82s/it]
 43%|████▎     | 6292/14725 [3:13:38<4:13:58,  1.81s/it]
 43%|████▎     | 6293/14725 [3:13:40<4:11:43,  1.79s/it]
 43%|████▎     | 6294/14725 [3:13:42<4:10:19,  1.78s/it]
 43%|████▎     | 6295/14725 [3:13:44<4:10:36,  1.78s/it]
 43%|████▎     | 6296/14725 [3:13:46<4:09:43,  1.78s/it]
 43%|████▎     | 6297/14725 [3:13:47<4:15:20,  1.82s/it]
 43%|████▎     | 6298/14725 [3:13:49<4:15:14,  1.82s/it]
 43%|████▎     | 6299/14725 [3:13:51<4:15:21,  1.82s/it]
 43%|████▎     | 6300/14725 [3:13:53<4:12:40,  1.80s/it]
                                                        

 43%|████▎     | 6300/14725 [3:13:53<4:12:40,  1.80s/it]
 43%|████▎     | 6301/14725 [3:13:55<4:20:22,  1.85s/it]
 43%|████▎     | 6302/14725 [3:13:57<4:19:01,  1.85s/
+0: {'loss': 0.8208, 'grad_norm': 0.7999156207955255, 'learning_rate': 2e-05, 'epoch': 0.43}
+0: it]
 43%|████▎     | 6303/14725 [3:13:58<4:17:54,  1.84s/it]
 43%|████▎     | 6304/14725 [3:14:00<4:16:33,  1.83s/it]
 43%|████▎     | 6305/14725 [3:14:02<4:17:39,  1.84s/it]
 43%|████▎     | 6306/14725 [3:14:04<4:17:41,  1.84s/it]
 43%|████▎     | 6307/14725 [3:14:06<4:14:14,  1.81s/it]
 43%|████▎     | 6308/14725 [3:14:08<4:15:37,  1.82s/it]
 43%|████▎     | 6309/14725 [3:14:09<4:13:08,  1.80s/it]
 43%|████▎     | 6310/14725 [3:14:11<4:13:50,  1.81s/it]
                                                        

 43%|████▎     | 6310/14725 [3:14:11<4:13:50,  1.81s/it]
 43%|████▎     | 6311/14725 [3:14:13<4:12:31,  1.80s/it]
 43%|████▎     | 6312/14725 [3:14:15<4:10:50,  1.79s/it]
 43%|████▎     | 6313/14725 [3:14:16<4:12:33,  1.80s/it]
 43%|████▎     | 6314/14725 [3:14:18<4:12:11,  1.80s/it]
 43%|████▎     | 6315/14725 [3:14:20<4:11:14,  1.79s/it]
 43%|████▎    
+0: {'loss': 0.8398, 'grad_norm': 0.7945636174675489, 'learning_rate': 2e-05, 'epoch': 0.43}
+0:  | 6316/14725 [3:14:22<4:09:37,  1.78s/it]
 43%|████▎     | 6317/14725 [3:14:24<4:10:53,  1.79s/it]
 43%|████▎     | 6318/14725 [3:14:25<4:12:02,  1.80s/it]
 43%|████▎     | 6319/14725 [3:14:27<4:11:58,  1.80s/it]
 43%|████▎     | 6320/14725 [3:14:29<4:11:19,  1.79s/it]
                                                        

 43%|████▎     | 6320/14725 [3:14:29<4:11:19,  1.79s/it]
 43%|████▎     | 6321/14725 [3:14:31<4:10:06,  1.79s/it]
 43%|████▎     | 6322/14725 [3:14:33<4:09:37,  1.78s/it]
 43%|████▎     | 6323/14725 [3:14:34<4:12:19,  1.80s/it]
 43%|████▎     | 6324/14725 [3:14:36<4:10:07,  1.79s/it]
 43%|████▎     | 6325/14725 [3:14:38<4:09:03,  1.78s/it]
 43%|████▎     | 6326/14725 [3:14:40<4:09:28,  1.78s/it]
 43%|████▎     | 6327/14725 [3:14:41<4:09:09,  1.78s/it]
 43%|████▎     | 6328/14725 [3:14:43<4:09:56,  1.79s/it]
 43%|████▎     | 6329/14725 [3:14:45<4:09:
+0: {'loss': 0.8432, 'grad_norm': 0.727613902224427, 'learning_rate': 2e-05, 'epoch': 0.43}
+0: {'loss': 0.8094, 'grad_norm': 0.7706756741585659, 'learning_rate': 2e-05, 'epoch': 0.43}
+0: 01,  1.78s/it]
 43%|████▎     | 6330/14725 [3:14:47<4:08:20,  1.77s/it]
                                                        

 43%|████▎     | 6330/14725 [3:14:47<4:08:20,  1.77s/it]
 43%|████▎     | 6331/14725 [3:14:49<4:11:59,  1.80s/it]
 43%|████▎     | 6332/14725 [3:14:50<4:11:32,  1.80s/it]
 43%|████▎     | 6333/14725 [3:14:52<4:11:12,  1.80s/it]
 43%|████▎     | 6334/14725 [3:14:54<4:14:19,  1.82s/it]
 43%|████▎     | 6335/14725 [3:14:56<4:13:07,  1.81s/it]
 43%|████▎     | 6336/14725 [3:14:58<4:13:00,  1.81s/it]
 43%|████▎     | 6337/14725 [3:15:00<4:12:28,  1.81s/it]
 43%|████▎     | 6338/14725 [3:15:01<4:11:53,  1.80s/it]
 43%|████▎     | 6339/14725 [3:15:03<4:10:31,  1.79s/it]
 43%|████▎     | 6340/14725 [3:15:05<4:13:26,  1.81s/it]
                                                        

 43%|████▎     | 6340/14725 [3:15:05<4:13:26,  1.81s/it]
 43%|████▎  
+0: {'loss': 0.834, 'grad_norm': 0.7029710696877144, 'learning_rate': 2e-05, 'epoch': 0.43}
+0:    | 6341/14725 [3:15:07<4:13:04,  1.81s/it]
 43%|████▎     | 6342/14725 [3:15:09<4:11:27,  1.80s/it]
 43%|████▎     | 6343/14725 [3:15:10<4:13:42,  1.82s/it]
 43%|████▎     | 6344/14725 [3:15:12<4:15:11,  1.83s/it]
 43%|████▎     | 6345/14725 [3:15:14<4:17:06,  1.84s/it]
 43%|████▎     | 6346/14725 [3:15:16<4:16:29,  1.84s/it]
 43%|████▎     | 6347/14725 [3:15:18<4:16:42,  1.84s/it]
 43%|████▎     | 6348/14725 [3:15:20<4:16:27,  1.84s/it]
 43%|████▎     | 6349/14725 [3:15:22<4:18:51,  1.85s/it]
 43%|████▎     | 6350/14725 [3:15:23<4:16:28,  1.84s/it]
                                                        

 43%|████▎     | 6350/14725 [3:15:23<4:16:28,  1.84s/it]
 43%|████▎     | 6351/14725 [3:15:25<4:15:15,  1.83s/it]
 43%|████▎     | 6352/14725 [3:15:27<4:13:57,  1.82s/it]
 43%|████▎     | 6353/14725 [3:15:29<4:13:01,  1.81s/it]
 43%|████▎     | 6354/14725 [3:15:31<4:1
+0: {'loss': 0.8212, 'grad_norm': 0.7356020825014268, 'learning_rate': 2e-05, 'epoch': 0.43}
+0: 3:36,  1.82s/it]
 43%|████▎     | 6355/14725 [3:15:32<4:11:02,  1.80s/it]
 43%|████▎     | 6356/14725 [3:15:34<4:09:53,  1.79s/it]
 43%|████▎     | 6357/14725 [3:15:36<4:08:12,  1.78s/it]
 43%|████▎     | 6358/14725 [3:15:38<4:08:43,  1.78s/it]
 43%|████▎     | 6359/14725 [3:15:39<4:08:43,  1.78s/it]
 43%|████▎     | 6360/14725 [3:15:41<4:07:38,  1.78s/it]
                                                        

 43%|████▎     | 6360/14725 [3:15:41<4:07:38,  1.78s/it]
 43%|████▎     | 6361/14725 [3:15:43<4:08:43,  1.78s/it]
 43%|████▎     | 6362/14725 [3:15:45<4:09:32,  1.79s/it]
 43%|████▎     | 6363/14725 [3:15:47<4:09:43,  1.79s/it]
 43%|████▎     | 6364/14725 [3:15:48<4:14:19,  1.83s/it]
 43%|████▎     | 6365/14725 [3:15:50<4:12:19,  1.81s/it]
 43%|████▎     | 6366/14725 [3:15:52<4:15:50,  1.84s/it]
 43%|████▎     | 6367/14725 [3:15:54<4:18:49,  1.86s/it]
 43%|██
+0: {'loss': 0.8292, 'grad_norm': 0.8137870351341087, 'learning_rate': 2e-05, 'epoch': 0.43}
+0: ██▎     | 6368/14725 [3:15:56<4:16:11,  1.84s/it]
 43%|████▎     | 6369/14725 [3:15:58<4:14:08,  1.82s/it]
 43%|████▎     | 6370/14725 [3:15:59<4:11:22,  1.81s/it]
                                                        

 43%|████▎     | 6370/14725 [3:15:59<4:11:22,  1.81s/it]
 43%|████▎     | 6371/14725 [3:16:01<4:10:23,  1.80s/it]
 43%|████▎     | 6372/14725 [3:16:03<4:10:26,  1.80s/it]
 43%|████▎     | 6373/14725 [3:16:05<4:11:34,  1.81s/it]
 43%|████▎     | 6374/14725 [3:16:07<4:09:43,  1.79s/it]
 43%|████▎     | 6375/14725 [3:16:08<4:08:52,  1.79s/it]
 43%|████▎     | 6376/14725 [3:16:10<4:10:26,  1.80s/it]
 43%|████▎     | 6377/14725 [3:16:12<4:11:35,  1.81s/it]
 43%|████▎     | 6378/14725 [3:16:14<4:13:02,  1.82s/it]
 43%|████▎     | 6379/14725 [3:16:16<4:12:48,  1.82s/it]
 43%|████▎     | 6380/14725 [3:16:18<4:27:03,  1.92s/it]
                                       
+0: {'loss': 0.8255, 'grad_norm': 0.8401089274326893, 'learning_rate': 2e-05, 'epoch': 0.43}
+0: {'loss': 0.8257, 'grad_norm': 0.7620659897623538, 'learning_rate': 2e-05, 'epoch': 0.43}
+0:                  

 43%|████▎     | 6380/14725 [3:16:18<4:27:03,  1.92s/it]
 43%|████▎     | 6381/14725 [3:16:20<4:25:18,  1.91s/it]
 43%|████▎     | 6382/14725 [3:16:22<4:23:32,  1.90s/it]
 43%|████▎     | 6383/14725 [3:16:23<4:17:54,  1.86s/it]
 43%|████▎     | 6384/14725 [3:16:25<4:15:28,  1.84s/it]
 43%|████▎     | 6385/14725 [3:16:27<4:15:24,  1.84s/it]
 43%|████▎     | 6386/14725 [3:16:29<4:17:07,  1.85s/it]
 43%|████▎     | 6387/14725 [3:16:31<4:13:48,  1.83s/it]
 43%|████▎     | 6388/14725 [3:16:33<4:19:19,  1.87s/it]
 43%|████▎     | 6389/14725 [3:16:35<4:23:19,  1.90s/it]
 43%|████▎     | 6390/14725 [3:16:36<4:26:21,  1.92s/it]
                                                        

 43%|████▎     | 6390/14725 [3:16:37<4:26:21,  1.92s/it]
 43%|████▎     | 6391/14725 [3:16:38<4:22:55,  1.89s/it]
 43%|████▎     | 6392/14725 [3:16:40<4:24:34,  1.91s/it]
 43%|█�
+0: {'loss': 0.8307, 'grad_norm': 0.812800757265187, 'learning_rate': 2e-05, 'epoch': 0.43}
+0: ��██▎     | 6393/14725 [3:16:42<4:22:00,  1.89s/it]
 43%|████▎     | 6394/14725 [3:16:44<4:21:53,  1.89s/it]
 43%|████▎     | 6395/14725 [3:16:46<4:19:32,  1.87s/it]
 43%|████▎     | 6396/14725 [3:16:48<4:19:42,  1.87s/it]
 43%|████▎     | 6397/14725 [3:16:49<4:16:33,  1.85s/it]
 43%|████▎     | 6398/14725 [3:16:51<4:13:20,  1.83s/it]
 43%|████▎     | 6399/14725 [3:16:53<4:10:22,  1.80s/it]
 43%|████▎     | 6400/14725 [3:16:55<4:10:43,  1.81s/it]
                                                        

 43%|████▎     | 6400/14725 [3:16:55<4:10:43,  1.81s/it]
 43%|████▎     | 6401/14725 [3:16:57<4:09:30,  1.80s/it]
 43%|████▎     | 6402/14725 [3:16:58<4:12:24,  1.82s/it]
 43%|████▎     | 6403/14725 [3:17:00<4:09:48,  1.80s/it]
 43%|████▎     | 6404/14725 [3:17:02<4:09:07,  1.80s/it]
 43%|████▎     | 6405/14725 [3:17:04<4:15:36,  1.84s/it]
 44%|████▎     | 6406/14725
+0: {'loss': 0.8285, 'grad_norm': 0.7624316377469861, 'learning_rate': 2e-05, 'epoch': 0.44}
+0:  [3:17:06<4:12:33,  1.82s/it]
 44%|████▎     | 6407/14725 [3:17:08<4:10:59,  1.81s/it]
 44%|████▎     | 6408/14725 [3:17:09<4:12:06,  1.82s/it]
 44%|████▎     | 6409/14725 [3:17:11<4:11:52,  1.82s/it]
 44%|████▎     | 6410/14725 [3:17:13<4:15:17,  1.84s/it]
                                                        

 44%|████▎     | 6410/14725 [3:17:13<4:15:17,  1.84s/it]
 44%|████▎     | 6411/14725 [3:17:15<4:12:21,  1.82s/it]
 44%|████▎     | 6412/14725 [3:17:17<4:14:23,  1.84s/it]
 44%|████▎     | 6413/14725 [3:17:19<4:17:52,  1.86s/it]
 44%|████▎     | 6414/14725 [3:17:20<4:13:37,  1.83s/it]
 44%|████▎     | 6415/14725 [3:17:22<4:15:00,  1.84s/it]
 44%|████▎     | 6416/14725 [3:17:24<4:13:19,  1.83s/it]
 44%|████▎     | 6417/14725 [3:17:26<4:15:31,  1.85s/it]
 44%|████▎     | 6418/14725 [3:17:28<4:11:29,  1.82s/it]
 44%|████▎     | 6419/14725 [3:17:30<4:13:48,  1.83s/it
+0: {'loss': 0.8225, 'grad_norm': 0.7002524786781243, 'learning_rate': 2e-05, 'epoch': 0.44}
+0: {'loss': 0.8128, 'grad_norm': 0.7646214852479469, 'learning_rate': 2e-05, 'epoch': 0.44}
+0: ]
 44%|████▎     | 6420/14725 [3:17:31<4:10:08,  1.81s/it]
                                                        

 44%|████▎     | 6420/14725 [3:17:31<4:10:08,  1.81s/it]
 44%|████▎     | 6421/14725 [3:17:33<4:12:51,  1.83s/it]
 44%|████▎     | 6422/14725 [3:17:35<4:10:05,  1.81s/it]
 44%|████▎     | 6423/14725 [3:17:37<4:08:19,  1.79s/it]
 44%|████▎     | 6424/14725 [3:17:39<4:08:11,  1.79s/it]
 44%|████▎     | 6425/14725 [3:17:40<4:08:00,  1.79s/it]
 44%|████▎     | 6426/14725 [3:17:42<4:07:55,  1.79s/it]
 44%|████▎     | 6427/14725 [3:17:44<4:10:41,  1.81s/it]
 44%|████▎     | 6428/14725 [3:17:46<4:10:55,  1.81s/it]
 44%|████▎     | 6429/14725 [3:17:48<4:08:23,  1.80s/it]
 44%|████▎     | 6430/14725 [3:17:49<4:09:05,  1.80s/it]
                                                        

 44%|████▎     | 6430/14725 [3:17:49<4:09:05,  1.80s/it]
 44%|████▎     | 6431/147
+0: {'loss': 0.8263, 'grad_norm': 0.7359245926053817, 'learning_rate': 2e-05, 'epoch': 0.44}
+0: 25 [3:17:51<4:09:42,  1.81s/it]
 44%|████▎     | 6432/14725 [3:17:53<4:11:12,  1.82s/it]
 44%|████▎     | 6433/14725 [3:17:55<4:09:21,  1.80s/it]
 44%|████▎     | 6434/14725 [3:17:57<4:08:33,  1.80s/it]
 44%|████▎     | 6435/14725 [3:17:58<4:08:29,  1.80s/it]
 44%|████▎     | 6436/14725 [3:18:00<4:11:05,  1.82s/it]
 44%|████▎     | 6437/14725 [3:18:02<4:11:14,  1.82s/it]
 44%|████▎     | 6438/14725 [3:18:04<4:09:46,  1.81s/it]
 44%|████▎     | 6439/14725 [3:18:06<4:09:22,  1.81s/it]
 44%|████▎     | 6440/14725 [3:18:07<4:11:05,  1.82s/it]
                                                        

 44%|████▎     | 6440/14725 [3:18:07<4:11:05,  1.82s/it]
 44%|████▎     | 6441/14725 [3:18:09<4:09:45,  1.81s/it]
 44%|████▎     | 6442/14725 [3:18:11<4:08:24,  1.80s/it]
 44%|████▍     | 6443/14725 [3:18:13<4:10:24,  1.81s/it]
 44%|████▍     | 6444/14725 [3:18:15<4:08:45,  1.80s/
+0: {'loss': 0.7983, 'grad_norm': 0.7088342089928569, 'learning_rate': 2e-05, 'epoch': 0.44}
+0: it]
 44%|████▍     | 6445/14725 [3:18:16<4:07:35,  1.79s/it]
 44%|████▍     | 6446/14725 [3:18:18<4:09:25,  1.81s/it]
 44%|████▍     | 6447/14725 [3:18:20<4:09:12,  1.81s/it]
 44%|████▍     | 6448/14725 [3:18:22<4:11:21,  1.82s/it]
 44%|████▍     | 6449/14725 [3:18:24<4:08:44,  1.80s/it]
 44%|████▍     | 6450/14725 [3:18:25<4:07:52,  1.80s/it]
                                                        

 44%|████▍     | 6450/14725 [3:18:25<4:07:52,  1.80s/it]
 44%|████▍     | 6451/14725 [3:18:27<4:07:50,  1.80s/it]
 44%|████▍     | 6452/14725 [3:18:29<4:12:32,  1.83s/it]
 44%|████▍     | 6453/14725 [3:18:31<4:14:59,  1.85s/it]
 44%|████▍     | 6454/14725 [3:18:33<4:17:23,  1.87s/it]
 44%|████▍     | 6455/14725 [3:18:35<4:13:16,  1.84s/it]
 44%|████▍     | 6456/14725 [3:18:37<4:11:51,  1.83s/it]
 44%|████▍     | 6457/14725 [3:18:38<4:11:38,  1.83s/it]
 44%|████▍    
+0: {'loss': 0.8271, 'grad_norm': 0.7537638858226099, 'learning_rate': 2e-05, 'epoch': 0.44}
+0:  | 6458/14725 [3:18:40<4:13:26,  1.84s/it]
 44%|████▍     | 6459/14725 [3:18:42<4:11:56,  1.83s/it]
 44%|████▍     | 6460/14725 [3:18:44<4:08:41,  1.81s/it]
                                                        

 44%|████▍     | 6460/14725 [3:18:44<4:08:41,  1.81s/it]
 44%|████▍     | 6461/14725 [3:18:46<4:09:58,  1.81s/it]
 44%|████▍     | 6462/14725 [3:18:47<4:07:53,  1.80s/it]
 44%|���███▍     | 6463/14725 [3:18:49<4:08:53,  1.81s/it]
 44%|████▍     | 6464/14725 [3:18:51<4:08:21,  1.80s/it]
 44%|████▍     | 6465/14725 [3:18:53<4:10:37,  1.82s/it]
 44%|████▍     | 6466/14725 [3:18:55<4:09:41,  1.81s/it]
 44%|████▍     | 6467/14725 [3:18:57<4:09:33,  1.81s/it]
 44%|████▍     | 6468/14725 [3:18:58<4:11:11,  1.83s/it]
 44%|████▍     | 6469/14725 [3:19:00<4:08:08,  1.80s/it]
 44%|████▍     | 6470/14725 [3:19:02<4:05:59,  1.79s/it]
                                                    
+0: {'loss': 0.8211, 'grad_norm': 0.7281232387282965, 'learning_rate': 2e-05, 'epoch': 0.44}
+0: {'loss': 0.8259, 'grad_norm': 0.7365370128071612, 'learning_rate': 2e-05, 'epoch': 0.44}
+0:     

 44%|████▍     | 6470/14725 [3:19:02<4:05:59,  1.79s/it]
 44%|████▍     | 6471/14725 [3:19:04<4:09:42,  1.82s/it]
 44%|████▍     | 6472/14725 [3:19:06<4:09:01,  1.81s/it]
 44%|████▍     | 6473/14725 [3:19:07<4:12:18,  1.83s/it]
 44%|████▍     | 6474/14725 [3:19:09<4:11:49,  1.83s/it]
 44%|████▍     | 6475/14725 [3:19:11<4:11:24,  1.83s/it]
 44%|████▍     | 6476/14725 [3:19:13<4:13:41,  1.85s/it]
 44%|████▍     | 6477/14725 [3:19:15<4:09:53,  1.82s/it]
 44%|████▍     | 6478/14725 [3:19:17<4:10:33,  1.82s/it]
 44%|████▍     | 6479/14725 [3:19:18<4:10:48,  1.82s/it]
 44%|████▍     | 6480/14725 [3:19:20<4:08:31,  1.81s/it]
                                                        

 44%|████▍     | 6480/14725 [3:19:20<4:08:31,  1.81s/it]
 44%|████▍     | 6481/14725 [3:19:22<4:09:56,  1.82s/it]
 44%|████▍     | 6482/14725 [3:19:24<4:11:07,  1.83s/it]
 44%|████▍  
+0: {'loss': 0.8341, 'grad_norm': 0.7991126157484738, 'learning_rate': 2e-05, 'epoch': 0.44}
+0:    | 6483/14725 [3:19:26<4:10:08,  1.82s/it]
 44%|████▍     | 6484/14725 [3:19:28<4:12:08,  1.84s/it]
 44%|████▍     | 6485/14725 [3:19:29<4:11:35,  1.83s/it]
 44%|████▍     | 6486/14725 [3:19:31<4:11:24,  1.83s/it]
 44%|████▍     | 6487/14725 [3:19:33<4:09:35,  1.82s/it]
 44%|████▍     | 6488/14725 [3:19:35<4:11:20,  1.83s/it]
 44%|████▍     | 6489/14725 [3:19:37<4:12:26,  1.84s/it]
 44%|████▍     | 6490/14725 [3:19:39<4:12:36,  1.84s/it]
                                                        

 44%|████▍     | 6490/14725 [3:19:39<4:12:36,  1.84s/it]
 44%|████▍     | 6491/14725 [3:19:40<4:10:02,  1.82s/it]
 44%|████▍     | 6492/14725 [3:19:42<4:11:49,  1.84s/it]
 44%|████▍     | 6493/14725 [3:19:44<4:10:04,  1.82s/it]
 44%|████▍     | 6494/14725 [3:19:46<4:09:25,  1.82s/it]
 44%|████▍     | 6495/14725 [3:19:48<4:13:57,  1.85s/it]
 44%|████▍     | 6496/14725 [3:19:50<4:1
+0: {'loss': 0.8306, 'grad_norm': 0.7591560347383, 'learning_rate': 2e-05, 'epoch': 0.44}
+0: 5:36,  1.86s/it]
 44%|████▍     | 6497/14725 [3:19:51<4:14:30,  1.86s/it]
 44%|████▍     | 6498/14725 [3:19:53<4:11:39,  1.84s/it]
 44%|████▍     | 6499/14725 [3:19:55<4:08:13,  1.81s/it]
 44%|████▍     | 6500/14725 [3:19:57<4:07:10,  1.80s/it]
                                                        

 44%|████▍     | 6500/14725 [3:19:57<4:07:10,  1.80s/it]
 44%|████▍     | 6501/14725 [3:19:59<4:07:44,  1.81s/it]
 44%|████▍     | 6502/14725 [3:20:00<4:08:15,  1.81s/it]
 44%|████▍     | 6503/14725 [3:20:02<4:06:18,  1.80s/it]
 44%|████▍     | 6504/14725 [3:20:04<4:04:55,  1.79s/it]
 44%|████▍     | 6505/14725 [3:20:06<4:04:49,  1.79s/it]
 44%|████▍     | 6506/14725 [3:20:08<4:06:50,  1.80s/it]
 44%|████▍     | 6507/14725 [3:20:09<4:07:03,  1.80s/it]
 44%|████▍     | 6508/14725 [3:20:11<4:10:21,  1.83s/it]
 44%|████▍     | 6509/14725 [3:20:13<4:10:41,  1.83s/it]
 44%|██
+0: {'loss': 0.8179, 'grad_norm': 0.7141771742612788, 'learning_rate': 2e-05, 'epoch': 0.44}
+0: {'loss': 0.832, 'grad_norm': 0.7254246396039574, 'learning_rate': 2e-05, 'epoch': 0.44}
+0: ██▍     | 6510/14725 [3:20:15<4:08:38,  1.82s/it]
                                                        

 44%|████▍     | 6510/14725 [3:20:15<4:08:38,  1.82s/it]
 44%|████▍     | 6511/14725 [3:20:17<4:11:04,  1.83s/it]
 44%|████▍     | 6512/14725 [3:20:19<4:12:45,  1.85s/it]
 44%|████▍     | 6513/14725 [3:20:20<4:11:53,  1.84s/it]
 44%|████▍     | 6514/14725 [3:20:22<4:14:44,  1.86s/it]
 44%|████▍     | 6515/14725 [3:20:24<4:13:44,  1.85s/it]
 44%|████▍     | 6516/14725 [3:20:26<4:11:16,  1.84s/it]
 44%|████▍     | 6517/14725 [3:20:28<4:09:18,  1.82s/it]
 44%|████▍     | 6518/14725 [3:20:30<4:09:13,  1.82s/it]
 44%|████▍     | 6519/14725 [3:20:31<4:05:43,  1.80s/it]
 44%|████▍     | 6520/14725 [3:20:33<4:08:48,  1.82s/it]
                                                        

 44%|████▍     | 6520/14725 [3:20:33<4:08:48,  1.82s/it]
 44%|████▍     | 6521/14725 [3:20:35<4
+0: {'loss': 0.8341, 'grad_norm': 0.7287272633522875, 'learning_rate': 2e-05, 'epoch': 0.44}
+0: :08:43,  1.82s/it]
 44%|████▍     | 6522/14725 [3:20:37<4:08:07,  1.81s/it]
 44%|████▍     | 6523/14725 [3:20:39<4:10:05,  1.83s/it]
 44%|████▍     | 6524/14725 [3:20:41<4:09:53,  1.83s/it]
 44%|████▍     | 6525/14725 [3:20:42<4:07:44,  1.81s/it]
 44%|████▍     | 6526/14725 [3:20:44<4:11:05,  1.84s/it]
 44%|████▍     | 6527/14725 [3:20:46<4:07:34,  1.81s/it]
 44%|████▍     | 6528/14725 [3:20:48<4:05:41,  1.80s/it]
 44%|████▍     | 6529/14725 [3:20:49<4:05:08,  1.79s/it]
 44%|████▍     | 6530/14725 [3:20:51<4:05:33,  1.80s/it]
                                                        

 44%|████▍     | 6530/14725 [3:20:51<4:05:33,  1.80s/it]
 44%|████▍     | 6531/14725 [3:20:53<4:06:47,  1.81s/it]
 44%|████▍     | 6532/14725 [3:20:55<4:08:24,  1.82s/it]
 44%|████▍     | 6533/14725 [3:20:57<4:06:45,  1.81s/it]
 44%|████▍     | 6534/14725 [3:20:59<4:11:28,  1.84s/it]
 44%|█�
+0: {'loss': 0.8335, 'grad_norm': 0.8025534816118918, 'learning_rate': 2e-05, 'epoch': 0.44}
+0: ��██▍     | 6535/14725 [3:21:00<4:07:25,  1.81s/it]
 44%|████▍     | 6536/14725 [3:21:02<4:08:18,  1.82s/it]
 44%|████▍     | 6537/14725 [3:21:04<4:05:52,  1.80s/it]
 44%|████▍     | 6538/14725 [3:21:06<4:08:45,  1.82s/it]
 44%|████▍     | 6539/14725 [3:21:08<4:09:31,  1.83s/it]
 44%|████▍     | 6540/14725 [3:21:10<4:09:22,  1.83s/it]
                                                        

 44%|████▍     | 6540/14725 [3:21:10<4:09:22,  1.83s/it]
 44%|████▍     | 6541/14725 [3:21:11<4:06:30,  1.81s/it]
 44%|████▍     | 6542/14725 [3:21:13<4:06:15,  1.81s/it]
 44%|████▍     | 6543/14725 [3:21:15<4:04:40,  1.79s/it]
 44%|████▍     | 6544/14725 [3:21:17<4:04:54,  1.80s/it]
 44%|████▍     | 6545/14725 [3:21:18<4:03:13,  1.78s/it]
 44%|████▍     | 6546/14725 [3:21:20<4:08:10,  1.82s/it]
 44%|████▍     | 6547/14725 [3:21:22<4:09:44,  1.83s/it]
 44%|████▍     | 6548/14725
+0: {'loss': 0.8046, 'grad_norm': 0.7899882332797601, 'learning_rate': 2e-05, 'epoch': 0.44}
+0: {'loss': 0.8146, 'grad_norm': 0.7169780838908134, 'learning_rate': 2e-05, 'epoch': 0.45}
+0:  [3:21:24<4:08:42,  1.82s/it]
 44%|████▍     | 6549/14725 [3:21:26<4:09:50,  1.83s/it]
 44%|████▍     | 6550/14725 [3:21:28<4:09:27,  1.83s/it]
                                                        

 44%|████▍     | 6550/14725 [3:21:28<4:09:27,  1.83s/it]
 44%|████▍     | 6551/14725 [3:21:30<4:13:37,  1.86s/it]
 44%|████▍     | 6552/14725 [3:21:32<4:14:15,  1.87s/it]
 45%|████▍     | 6553/14725 [3:21:33<4:11:20,  1.85s/it]
 45%|████▍     | 6554/14725 [3:21:35<4:08:21,  1.82s/it]
 45%|████▍     | 6555/14725 [3:21:37<4:09:12,  1.83s/it]
 45%|████▍     | 6556/14725 [3:21:39<4:09:01,  1.83s/it]
 45%|████▍     | 6557/14725 [3:21:40<4:06:02,  1.81s/it]
 45%|████▍     | 6558/14725 [3:21:42<4:05:09,  1.80s/it]
 45%|████▍     | 6559/14725 [3:21:44<4:03:13,  1.79s/it]
 45%|████▍     | 6560/14725 [3:21:46<4:02:40,  1.78s/it]
                                                        

 45%|�
+0: {'loss': 0.8124, 'grad_norm': 0.7431053710434535, 'learning_rate': 2e-05, 'epoch': 0.45}
+0: �███▍     | 6560/14725 [3:21:46<4:02:40,  1.78s/it]
 45%|████▍     | 6561/14725 [3:21:48<4:04:00,  1.79s/it]
 45%|████▍     | 6562/14725 [3:21:50<4:07:11,  1.82s/it]
 45%|████▍     | 6563/14725 [3:21:51<4:08:58,  1.83s/it]
 45%|████▍     | 6564/14725 [3:21:53<4:08:00,  1.82s/it]
 45%|████▍     | 6565/14725 [3:21:55<4:08:17,  1.83s/it]
 45%|████▍     | 6566/14725 [3:21:57<4:05:01,  1.80s/it]
 45%|████▍     | 6567/14725 [3:21:59<4:03:50,  1.79s/it]
 45%|████▍     | 6568/14725 [3:22:00<4:03:07,  1.79s/it]
 45%|████▍     | 6569/14725 [3:22:02<4:03:30,  1.79s/it]
 45%|████▍     | 6570/14725 [3:22:04<4:03:15,  1.79s/it]
                                                        

 45%|████▍     | 6570/14725 [3:22:04<4:03:15,  1.79s/it]
 45%|████▍     | 6571/14725 [3:22:06<4:03:59,  1.80s/it]
 45%|████▍     | 6572/14725 [3:22:07<4:04:07,  1.80s/it]
 45%|████▍     | 6573/147
+0: {'loss': 0.8214, 'grad_norm': 0.7576398441701423, 'learning_rate': 2e-05, 'epoch': 0.45}
+0: 25 [3:22:09<4:02:01,  1.78s/it]
 45%|████▍     | 6574/14725 [3:22:11<4:00:44,  1.77s/it]
 45%|████▍     | 6575/14725 [3:22:13<4:02:06,  1.78s/it]
 45%|████▍     | 6576/14725 [3:22:15<4:04:12,  1.80s/it]
 45%|████▍     | 6577/14725 [3:22:16<4:04:12,  1.80s/it]
 45%|█���██▍     | 6578/14725 [3:22:18<4:05:56,  1.81s/it]
 45%|████▍     | 6579/14725 [3:22:20<4:07:46,  1.83s/it]
 45%|████▍     | 6580/14725 [3:22:22<4:07:03,  1.82s/it]
                                                        

 45%|████▍     | 6580/14725 [3:22:22<4:07:03,  1.82s/it]
 45%|████▍     | 6581/14725 [3:22:24<4:06:40,  1.82s/it]
 45%|████▍     | 6582/14725 [3:22:26<4:07:09,  1.82s/it]
 45%|████▍     | 6583/14725 [3:22:27<4:08:02,  1.83s/it]
 45%|████▍     | 6584/14725 [3:22:29<4:09:19,  1.84s/it]
 45%|████▍     | 6585/14725 [3:22:31<4:09:58,  1.84s/it]
 45%|████▍     | 6586/14725 [3:22:33<4:11:00,  1.85s/
+0: {'loss': 0.8316, 'grad_norm': 0.6897426696107263, 'learning_rate': 2e-05, 'epoch': 0.45}
+0: it]
 45%|████▍     | 6587/14725 [3:22:35<4:08:56,  1.84s/it]
 45%|████▍     | 6588/14725 [3:22:37<4:09:26,  1.84s/it]
 45%|████▍     | 6589/14725 [3:22:38<4:07:02,  1.82s/it]
 45%|████▍     | 6590/14725 [3:22:40<4:05:46,  1.81s/it]
                                                        

 45%|████▍     | 6590/14725 [3:22:40<4:05:46,  1.81s/it]
 45%|████▍     | 6591/14725 [3:22:42<4:06:22,  1.82s/it]
 45%|████▍     | 6592/14725 [3:22:44<4:07:19,  1.82s/it]
 45%|████▍     | 6593/14725 [3:22:46<4:07:20,  1.82s/it]
 45%|████▍     | 6594/14725 [3:22:48<4:08:05,  1.83s/it]
 45%|████▍     | 6595/14725 [3:22:49<4:07:01,  1.82s/it]
 45%|████▍     | 6596/14725 [3:22:51<4:05:54,  1.82s/it]
 45%|████▍     | 6597/14725 [3:22:53<4:04:33,  1.81s/it]
 45%|████▍     | 6598/14725 [3:22:55<4:05:43,  1.81s/it]
 45%|████▍     | 6599/14725 [3:22:57<4:05:03,  1.81s/it]
 45%|████▍    
+0: {'loss': 0.8208, 'grad_norm': 0.7994283119506174, 'learning_rate': 2e-05, 'epoch': 0.45}
+0: {'loss': 0.8101, 'grad_norm': 0.7209094877963366, 'learning_rate': 2e-05, 'epoch': 0.45}
+0:  | 6600/14725 [3:22:58<4:06:58,  1.82s/it]
                                                        

 45%|████▍     | 6600/14725 [3:22:58<4:06:58,  1.82s/it]
 45%|████▍     | 6601/14725 [3:23:00<4:05:35,  1.81s/it]
 45%|████▍     | 6602/14725 [3:23:02<4:03:37,  1.80s/it]
 45%|████▍     | 6603/14725 [3:23:04<4:02:48,  1.79s/it]
 45%|████▍     | 6604/14725 [3:23:06<4:04:17,  1.80s/it]
 45%|████▍     | 6605/14725 [3:23:07<4:07:26,  1.83s/it]
 45%|████▍     | 6606/14725 [3:23:09<4:04:48,  1.81s/it]
 45%|████▍     | 6607/14725 [3:23:11<4:04:14,  1.81s/it]
 45%|████▍     | 6608/14725 [3:23:13<4:07:38,  1.83s/it]
 45%|████▍     | 6609/14725 [3:23:15<4:06:36,  1.82s/it]
 45%|████▍     | 6610/14725 [3:23:16<4:03:27,  1.80s/it]
                                                        

 45%|████▍     | 6610/14725 [3:23:16<4:03:27,  1.80s/it]
 45%|████▍     | 6611/14725 [3:23:18<4:03:25,  1.80
+0: {'loss': 0.8217, 'grad_norm': 0.7170952957164353, 'learning_rate': 2e-05, 'epoch': 0.45}
+0: s/it]
 45%|████▍     | 6612/14725 [3:23:20<4:06:00,  1.82s/it]
 45%|████▍     | 6613/14725 [3:23:22<4:05:57,  1.82s/it]
 45%|████▍     | 6614/14725 [3:23:24<4:04:06,  1.81s/it]
 45%|████▍     | 6615/14725 [3:23:26<4:04:43,  1.81s/it]
 45%|████▍     | 6616/14725 [3:23:27<4:03:57,  1.81s/it]
 45%|████▍     | 6617/14725 [3:23:29<4:02:26,  1.79s/it]
 45%|████▍     | 6618/14725 [3:23:31<4:03:24,  1.80s/it]
 45%|████▍     | 6619/14725 [3:23:33<4:04:22,  1.81s/it]
 45%|████▍     | 6620/14725 [3:23:35<4:03:36,  1.80s/it]
                                                        

 45%|████▍     | 6620/14725 [3:23:35<4:03:36,  1.80s/it]
 45%|████▍     | 6621/14725 [3:23:36<4:05:40,  1.82s/it]
 45%|████▍     | 6622/14725 [3:23:38<4:10:03,  1.85s/it]
 45%|████▍     | 6623/14725 [3:23:40<4:08:16,  1.84s/it]
 45%|████▍     | 6624/14725 [3:23:42<4:06:36,  1.83s/it]
 45%|████▍  
+0: {'loss': 0.8288, 'grad_norm': 0.7996032594026894, 'learning_rate': 2e-05, 'epoch': 0.45}
+0:    | 6625/14725 [3:23:44<4:10:06,  1.85s/it]
 45%|████▍     | 6626/14725 [3:23:46<4:06:44,  1.83s/it]
 45%|████▌     | 6627/14725 [3:23:47<4:06:01,  1.82s/it]
 45%|████▌     | 6628/14725 [3:23:49<4:08:56,  1.84s/it]
 45%|████▌     | 6629/14725 [3:23:51<4:09:27,  1.85s/it]
 45%|████▌     | 6630/14725 [3:23:53<4:06:28,  1.83s/it]
                                                        

 45%|████▌     | 6630/14725 [3:23:53<4:06:28,  1.83s/it]
 45%|████▌     | 6631/14725 [3:23:55<4:05:18,  1.82s/it]
 45%|████▌     | 6632/14725 [3:23:57<4:02:32,  1.80s/it]
 45%|████▌     | 6633/14725 [3:23:58<4:05:50,  1.82s/it]
 45%|████▌     | 6634/14725 [3:24:00<4:05:10,  1.82s/it]
 45%|████▌     | 6635/14725 [3:24:02<4:05:39,  1.82s/it]
 45%|████▌     | 6636/14725 [3:24:04<4:05:15,  1.82s/it]
 45%|████▌     | 6637/14725 [3:24:06<4:04:40,  1.82s/it]
 45%|████▌     | 6638/14725 [3:24:07<4:0
+0: {'loss': 0.819, 'grad_norm': 0.7553780689614351, 'learning_rate': 2e-05, 'epoch': 0.45}
+0: {'loss': 0.8724, 'grad_norm': 0.8034810214515635, 'learning_rate': 2e-05, 'epoch': 0.45}
+0: 3:37,  1.81s/it]
 45%|████▌     | 6639/14725 [3:24:09<4:01:54,  1.80s/it]
 45%|████▌     | 6640/14725 [3:24:11<4:01:19,  1.79s/it]
                                                        

 45%|████▌     | 6640/14725 [3:24:11<4:01:19,  1.79s/it]
 45%|████▌     | 6641/14725 [3:24:13<4:02:56,  1.80s/it]
 45%|████▌     | 6642/14725 [3:24:15<4:03:08,  1.80s/it]
 45%|████▌     | 6643/14725 [3:24:17<4:05:33,  1.82s/it]
 45%|████▌     | 6644/14725 [3:24:18<4:07:00,  1.83s/it]
 45%|████▌     | 6645/14725 [3:24:20<4:04:11,  1.81s/it]
 45%|████▌     | 6646/14725 [3:24:22<4:02:19,  1.80s/it]
 45%|████▌     | 6647/14725 [3:24:24<4:06:47,  1.83s/it]
 45%|████▌     | 6648/14725 [3:24:26<4:04:05,  1.81s/it]
 45%|████▌     | 6649/14725 [3:24:27<4:02:41,  1.80s/it]
 45%|████▌     | 6650/14725 [3:24:29<4:04:30,  1.82s/it]
                                                        

 45%|████▌
+0: {'loss': 0.8127, 'grad_norm': 0.7005391258173583, 'learning_rate': 2e-05, 'epoch': 0.45}
+0:      | 6650/14725 [3:24:29<4:04:30,  1.82s/it]
 45%|████▌     | 6651/14725 [3:24:31<4:03:39,  1.81s/it]
 45%|████▌     | 6652/14725 [3:24:33<4:03:10,  1.81s/it]
 45%|████▌     | 6653/14725 [3:24:35<4:02:14,  1.80s/it]
 45%|████▌     | 6654/14725 [3:24:36<3:59:58,  1.78s/it]
 45%|████▌     | 6655/14725 [3:24:38<3:58:36,  1.77s/it]
 45%|████▌     | 6656/14725 [3:24:40<3:58:31,  1.77s/it]
 45%|████▌     | 6657/14725 [3:24:42<3:57:55,  1.77s/it]
 45%|████▌     | 6658/14725 [3:24:43<4:01:17,  1.79s/it]
 45%|████▌     | 6659/14725 [3:24:45<4:01:36,  1.80s/it]
 45%|████▌     | 6660/14725 [3:24:47<4:00:19,  1.79s/it]
                                                        

 45%|████▌     | 6660/14725 [3:24:47<4:00:19,  1.79s/it]
 45%|████▌     | 6661/14725 [3:24:49<4:05:20,  1.83s/it]
 45%|████▌     | 6662/14725 [3:24:51<4:05:16,  1.83s/it]
 45%|████▌     | 6663/14725 [3:24:53<4
+0: {'loss': 0.8328, 'grad_norm': 0.7222007810595242, 'learning_rate': 2e-05, 'epoch': 0.45}
+0: :05:04,  1.82s/it]
 45%|████▌     | 6664/14725 [3:24:54<4:02:32,  1.81s/it]
 45%|████▌     | 6665/14725 [3:24:56<4:01:01,  1.79s/it]
 45%|████▌     | 6666/14725 [3:24:58<3:59:37,  1.78s/it]
 45%|████▌     | 6667/14725 [3:25:00<3:58:27,  1.78s/it]
 45%|████▌     | 6668/14725 [3:25:01<3:57:37,  1.77s/it]
 45%|████▌     | 6669/14725 [3:25:03<4:01:05,  1.80s/it]
 45%|████▌     | 6670/14725 [3:25:05<3:59:16,  1.78s/it]
                                                        

 45%|████▌     | 6670/14725 [3:25:05<3:59:16,  1.78s/it]
 45%|████▌     | 6671/14725 [3:25:07<4:02:09,  1.80s/it]
 45%|████▌     | 6672/14725 [3:25:09<4:02:37,  1.81s/it]
 45%|████▌     | 6673/14725 [3:25:10<4:01:37,  1.80s/it]
 45%|████▌     | 6674/14725 [3:25:12<4:00:37,  1.79s/it]
 45%|████▌     | 6675/14725 [3:25:14<4:05:34,  1.83s/it]
 45%|████▌     | 6676/14725 [3:25:16<4:05:35,  1.83s/it]
 45%|█�
+0: {'loss': 0.8134, 'grad_norm': 0.7111229968697869, 'learning_rate': 2e-05, 'epoch': 0.45}
+0: ��██▌     | 6677/14725 [3:25:18<4:05:39,  1.83s/it]
 45%|████▌     | 6678/14725 [3:25:20<4:02:06,  1.81s/it]
 45%|████▌     | 6679/14725 [3:25:21<4:00:06,  1.79s/it]
 45%|████▌     | 6680/14725 [3:25:23<3:59:13,  1.78s/it]
                                                        

 45%|████▌     | 6680/14725 [3:25:23<3:59:13,  1.78s/it]
 45%|████▌     | 6681/14725 [3:25:25<3:59:11,  1.78s/it]
 45%|████▌     | 6682/14725 [3:25:27<3:57:52,  1.77s/it]
 45%|████▌     | 6683/14725 [3:25:29<4:05:47,  1.83s/it]
 45%|████▌     | 6684/14725 [3:25:30<4:07:15,  1.84s/it]
 45%|████▌     | 6685/14725 [3:25:32<4:03:23,  1.82s/it]
 45%|████▌     | 6686/14725 [3:25:34<4:02:47,  1.81s/it]
 45%|████▌     | 6687/14725 [3:25:36<4:03:26,  1.82s/it]
 45%|████▌     | 6688/14725 [3:25:38<4:02:39,  1.81s/it]
 45%|████▌     | 6689/14725 [3:25:40<4:05:29,  1.83s/it]
 45%|████▌     | 6690/14725
+0: {'loss': 0.8202, 'grad_norm': 0.7167224799307871, 'learning_rate': 2e-05, 'epoch': 0.45}
+0: {'loss': 0.8468, 'grad_norm': 0.8330462308401795, 'learning_rate': 2e-05, 'epoch': 0.46}
+0:  [3:25:41<4:07:00,  1.84s/it]
                                                        

 45%|████▌     | 6690/14725 [3:25:41<4:07:00,  1.84s/it]
 45%|████���     | 6691/14725 [3:25:43<4:07:05,  1.85s/it]
 45%|████▌     | 6692/14725 [3:25:45<4:05:49,  1.84s/it]
 45%|████▌     | 6693/14725 [3:25:47<4:04:50,  1.83s/it]
 45%|████▌     | 6694/14725 [3:25:49<4:05:35,  1.83s/it]
 45%|████▌     | 6695/14725 [3:25:51<4:03:36,  1.82s/it]
 45%|████▌     | 6696/14725 [3:25:52<4:05:18,  1.83s/it]
 45%|████▌     | 6697/14725 [3:25:54<4:05:36,  1.84s/it]
 45%|████▌     | 6698/14725 [3:25:56<4:03:24,  1.82s/it]
 45%|████▌     | 6699/14725 [3:25:58<4:02:58,  1.82s/it]
 46%|████▌     | 6700/14725 [3:26:00<4:01:40,  1.81s/it]
                                                        

 46%|████▌     | 6700/14725 [3:26:00<4:01:40,  1.81s/it]
 46%|████▌     | 6701/14725 [3:26:01<4:01:09,  1.80s/it]
 46%|�
+0: {'loss': 0.8146, 'grad_norm': 0.7147695698543383, 'learning_rate': 2e-05, 'epoch': 0.46}
+0: �███▌     | 6702/14725 [3:26:03<4:00:07,  1.80s/it]
 46%|████▌     | 6703/14725 [3:26:05<3:59:03,  1.79s/it]
 46%|████▌     | 6704/14725 [3:26:07<4:01:38,  1.81s/it]
 46%|████▌     | 6705/14725 [3:26:09<4:00:43,  1.80s/it]
 46%|████▌     | 6706/14725 [3:26:10<4:00:36,  1.80s/it]
 46%|████▌     | 6707/14725 [3:26:12<3:58:05,  1.78s/it]
 46%|████▌     | 6708/14725 [3:26:14<3:59:58,  1.80s/it]
 46%|████▌     | 6709/14725 [3:26:16<3:59:58,  1.80s/it]
 46%|████▌     | 6710/14725 [3:26:18<4:03:29,  1.82s/it]
                                                        

 46%|████▌     | 6710/14725 [3:26:18<4:03:29,  1.82s/it]
 46%|████▌     | 6711/14725 [3:26:19<4:02:59,  1.82s/it]
 46%|████▌     | 6712/14725 [3:26:21<4:02:14,  1.81s/it]
 46%|████▌     | 6713/14725 [3:26:23<4:01:39,  1.81s/it]
 46%|████▌     | 6714/14725 [3:26:25<4:00:43,  1.80s/it]
 46%|████▌     | 6715/147
+0: {'loss': 0.8395, 'grad_norm': 0.7171797618098079, 'learning_rate': 2e-05, 'epoch': 0.46}
+0: 25 [3:26:27<3:59:43,  1.80s/it]
 46%|████▌     | 6716/14725 [3:26:28<3:59:30,  1.79s/it]
 46%|████▌     | 6717/14725 [3:26:30<3:59:30,  1.79s/it]
 46%|████▌     | 6718/14725 [3:26:32<3:59:26,  1.79s/it]
 46%|████▌     | 6719/14725 [3:26:34<4:01:12,  1.81s/it]
 46%|████▌     | 6720/14725 [3:26:36<4:01:53,  1.81s/it]
                                                        

 46%|████▌     | 6720/14725 [3:26:36<4:01:53,  1.81s/it]
 46%|████▌     | 6721/14725 [3:26:37<4:01:54,  1.81s/it]
 46%|████▌     | 6722/14725 [3:26:39<4:01:06,  1.81s/it]
 46%|████▌     | 6723/14725 [3:26:41<4:02:12,  1.82s/it]
 46%|████▌     | 6724/14725 [3:26:43<4:01:12,  1.81s/it]
 46%|████▌     | 6725/14725 [3:26:45<4:01:56,  1.81s/it]
 46%|████▌     | 6726/14725 [3:26:47<4:01:48,  1.81s/it]
 46%|████▌     | 6727/14725 [3:26:48<4:01:42,  1.81s/it]
 46%|████▌     | 6728/14725 [3:26:50<4:03:01,  1.82s/
+0: {'loss': 0.8244, 'grad_norm': 0.7974870682714389, 'learning_rate': 2e-05, 'epoch': 0.46}
+0: {'loss': 0.8131, 'grad_norm': 0.6682395810759395, 'learning_rate': 2e-05, 'epoch': 0.46}
+0: it]
 46%|████▌     | 6729/14725 [3:26:52<4:02:23,  1.82s/it]
 46%|████▌     | 6730/14725 [3:26:54<4:02:29,  1.82s/it]
                                                        

 46%|████▌     | 6730/14725 [3:26:54<4:02:29,  1.82s/it]
 46%|████▌     | 6731/14725 [3:26:56<4:03:18,  1.83s/it]
 46%|████▌     | 6732/14725 [3:26:57<4:03:36,  1.83s/it]
 46%|████▌     | 6733/14725 [3:26:59<4:03:27,  1.83s/it]
 46%|████▌     | 6734/14725 [3:27:01<4:03:26,  1.83s/it]
 46%|████▌     | 6735/14725 [3:27:03<4:04:15,  1.83s/it]
 46%|████▌     | 6736/14725 [3:27:05<4:08:22,  1.87s/it]
 46%|████▌     | 6737/14725 [3:27:07<4:06:56,  1.85s/it]
 46%|████▌     | 6738/14725 [3:27:09<4:03:55,  1.83s/it]
 46%|████▌     | 6739/14725 [3:27:10<4:00:40,  1.81s/it]
 46%|████▌     | 6740/14725 [3:27:12<4:01:01,  1.81s/it]
                                                        

 46%|████▌     | 6740/1
+0: {'loss': 0.8246, 'grad_norm': 0.7592167052934509, 'learning_rate': 2e-05, 'epoch': 0.46}
+0: 4725 [3:27:12<4:01:01,  1.81s/it]
 46%|████▌     | 6741/14725 [3:27:14<4:00:35,  1.81s/it]
 46%|████▌     | 6742/14725 [3:27:16<4:02:27,  1.82s/it]
 46%|████▌     | 6743/14725 [3:27:18<3:59:56,  1.80s/it]
 46%|████▌     | 6744/14725 [3:27:19<4:01:36,  1.82s/it]
 46%|████▌     | 6745/14725 [3:27:21<4:01:30,  1.82s/it]
 46%|████▌     | 6746/14725 [3:27:23<4:02:29,  1.82s/it]
 46%|████▌     | 6747/14725 [3:27:25<4:00:12,  1.81s/it]
 46%|████▌     | 6748/14725 [3:27:27<4:01:53,  1.82s/it]
 46%|████▌     | 6749/14725 [3:27:28<4:01:55,  1.82s/it]
 46%|████▌     | 6750/14725 [3:27:30<4:04:51,  1.84s/it]
                                                        

 46%|████▌     | 6750/14725 [3:27:30<4:04:51,  1.84s/it]
 46%|████▌     | 6751/14725 [3:27:32<4:03:50,  1.83s/it]
 46%|████▌     | 6752/14725 [3:27:34<4:03:04,  1.83s/it]
 46%|████▌     | 6753/14725 [3:27:36<4:00:36,  1.81
+0: {'loss': 0.8226, 'grad_norm': 0.7298482542675314, 'learning_rate': 2e-05, 'epoch': 0.46}
+0: s/it]
 46%|████▌     | 6754/14725 [3:27:38<3:58:51,  1.80s/it]
 46%|████▌     | 6755/14725 [3:27:39<3:57:43,  1.79s/it]
 46%|████▌     | 6756/14725 [3:27:41<4:00:55,  1.81s/it]
 46%|████▌     | 6757/14725 [3:27:43<4:01:13,  1.82s/it]
 46%|████▌     | 6758/14725 [3:27:45<3:59:29,  1.80s/it]
 46%|████▌     | 6759/14725 [3:27:47<4:02:47,  1.83s/it]
 46%|████▌     | 6760/14725 [3:27:48<3:59:26,  1.80s/it]
                                                        

 46%|████▌     | 6760/14725 [3:27:48<3:59:26,  1.80s/it]
 46%|████▌     | 6761/14725 [3:27:50<3:57:43,  1.79s/it]
 46%|████▌     | 6762/14725 [3:27:52<3:58:35,  1.80s/it]
 46%|████▌     | 6763/14725 [3:27:54<3:58:30,  1.80s/it]
 46%|████▌     | 6764/14725 [3:27:56<4:05:46,  1.85s/it]
 46%|████▌     | 6765/14725 [3:27:58<4:06:19,  1.86s/it]
 46%|████▌     | 6766/14725 [3:27:59<4:05:01,  1.85s/it]
 46%|████▌  
+0: {'loss': 0.8465, 'grad_norm': 0.726525603042793, 'learning_rate': 2e-05, 'epoch': 0.46}
+0:    | 6767/14725 [3:28:01<4:06:50,  1.86s/it]
 46%|████▌     | 6768/14725 [3:28:03<4:01:55,  1.82s/it]
 46%|████▌     | 6769/14725 [3:28:05<4:00:12,  1.81s/it]
 46%|████▌     | 6770/14725 [3:28:07<3:58:22,  1.80s/it]
                                                        

 46%|████▌     | 6770/14725 [3:28:07<3:58:22,  1.80s/it]
 46%|████▌     | 6771/14725 [3:28:08<3:59:06,  1.80s/it]
 46%|████▌     | 6772/14725 [3:28:10<3:57:44,  1.79s/it]
 46%|████▌     | 6773/14725 [3:28:12<4:03:51,  1.84s/it]
 46%|████▌     | 6774/14725 [3:28:14<4:02:44,  1.83s/it]
 46%|████▌     | 6775/14725 [3:28:16<4:04:30,  1.85s/it]
 46%|████▌     | 6776/14725 [3:28:18<4:05:38,  1.85s/it]
 46%|████▌     | 6777/14725 [3:28:19<4:02:06,  1.83s/it]
 46%|████▌     | 6778/14725 [3:28:21<3:59:45,  1.81s/it]
 46%|████▌     | 6779/14725 [3:28:23<3:59:16,  1.81s/it]
 46%|████▌     | 6780/14725 [3:28:25<3:5
+0: {'loss': 0.8247, 'grad_norm': 0.7644880566692392, 'learning_rate': 2e-05, 'epoch': 0.46}
+0: {'loss': 0.8113, 'grad_norm': 0.7237751742980173, 'learning_rate': 2e-05, 'epoch': 0.46}
+0: 6:44,  1.79s/it]
                                                        

 46%|████▌     | 6780/14725 [3:28:25<3:56:44,  1.79s/it]
 46%|████▌     | 6781/14725 [3:28:27<3:56:21,  1.79s/it]
 46%|████▌     | 6782/14725 [3:28:28<3:59:21,  1.81s/it]
 46%|████▌     | 6783/14725 [3:28:30<4:01:17,  1.82s/it]
 46%|████▌     | 6784/14725 [3:28:32<3:59:35,  1.81s/it]
 46%|████▌     | 6785/14725 [3:28:34<3:57:42,  1.80s/it]
 46%|████▌     | 6786/14725 [3:28:36<3:57:15,  1.79s/it]
 46%|████▌     | 6787/14725 [3:28:37<3:55:31,  1.78s/it]
 46%|████▌     | 6788/14725 [3:28:39<3:55:08,  1.78s/it]
 46%|████▌     | 6789/14725 [3:28:41<3:58:19,  1.80s/it]
 46%|████▌     | 6790/14725 [3:28:43<4:02:11,  1.83s/it]
                                                        

 46%|████▌     | 6790/14725 [3:28:43<4:02:11,  1.83s/it]
 46%|████▌     | 6791/14725 [3:28:45<4:03:16,  1.84s/it]
 46%|████▌
+0: {'loss': 0.835, 'grad_norm': 0.7122985269904221, 'learning_rate': 2e-05, 'epoch': 0.46}
+0:      | 6792/14725 [3:28:47<4:03:48,  1.84s/it]
 46%|████▌     | 6793/14725 [3:28:48<4:00:47,  1.82s/it]
 46%|████▌     | 6794/14725 [3:28:50<4:04:04,  1.85s/it]
 46%|████▌     | 6795/14725 [3:28:52<4:03:50,  1.85s/it]
 46%|████▌     | 6796/14725 [3:28:54<4:04:50,  1.85s/it]
 46%|████▌     | 6797/14725 [3:28:56<4:03:26,  1.84s/it]
 46%|████▌     | 6798/14725 [3:28:58<4:02:32,  1.84s/it]
 46%|████▌     | 6799/14725 [3:28:59<4:00:43,  1.82s/it]
 46%|████▌     | 6800/14725 [3:29:01<4:02:52,  1.84s/it]
                                                        

 46%|████▌     | 6800/14725 [3:29:01<4:02:52,  1.84s/it]
 46%|████▌     | 6801/14725 [3:29:03<4:02:18,  1.83s/it]
 46%|████▌     | 6802/14725 [3:29:05<4:00:08,  1.82s/it]
 46%|████▌     | 6803/14725 [3:29:07<4:00:33,  1.82s/it]
 46%|████▌     | 6804/14725 [3:29:09<3:59:22,  1.81s/it]
 46%|████▌     | 6805/14725 [3:29:10<3
+0: {'loss': 0.8176, 'grad_norm': 0.7360919202007141, 'learning_rate': 2e-05, 'epoch': 0.46}
+0: :57:46,  1.80s/it]
 46%|████▌     | 6806/14725 [3:29:12<3:58:24,  1.81s/it]
 46%|████▌     | 6807/14725 [3:29:14<3:56:02,  1.79s/it]
 46%|████▌     | 6808/14725 [3:29:16<3:57:05,  1.80s/it]
 46%|████▌     | 6809/14725 [3:29:17<3:57:18,  1.80s/it]
 46%|████▌     | 6810/14725 [3:29:19<3:55:18,  1.78s/it]
                                                        

 46%|████▌     | 6810/14725 [3:29:19<3:55:18,  1.78s/it]
 46%|████▋     | 6811/14725 [3:29:21<3:55:31,  1.79s/it]
 46%|████▋     | 6812/14725 [3:29:23<3:56:01,  1.79s/it]
 46%|████▋     | 6813/14725 [3:29:25<3:55:14,  1.78s/it]
 46%|████▋     | 6814/14725 [3:29:26<3:54:55,  1.78s/it]
 46%|████▋     | 6815/14725 [3:29:28<3:53:48,  1.77s/it]
 46%|████▋     | 6816/14725 [3:29:30<3:52:47,  1.77s/it]
 46%|████▋     | 6817/14725 [3:29:32<3:56:57,  1.80s/it]
 46%|████▋     | 6818/14725 [3:29:34<3:56:07,  1.79s/it]
 46%|█�
+0: {'loss': 0.8157, 'grad_norm': 0.7015126224601085, 'learning_rate': 2e-05, 'epoch': 0.46}
+0: {'loss': 0.8099, 'grad_norm': 0.7154004447346707, 'learning_rate': 2e-05, 'epoch': 0.46}
+0: ��██▋     | 6819/14725 [3:29:35<3:57:14,  1.80s/it]
 46%|████▋     | 6820/14725 [3:29:37<3:57:37,  1.80s/it]
                                                        

 46%|████▋     | 6820/14725 [3:29:37<3:57:37,  1.80s/it]
 46%|████▋     | 6821/14725 [3:29:39<3:57:10,  1.80s/it]
 46%|████▋     | 6822/14725 [3:29:41<3:57:20,  1.80s/it]
 46%|████▋     | 6823/14725 [3:29:43<3:57:08,  1.80s/it]
 46%|████▋     | 6824/14725 [3:29:44<4:02:21,  1.84s/it]
 46%|████▋     | 6825/14725 [3:29:46<4:00:51,  1.83s/it]
 46%|████▋     | 6826/14725 [3:29:48<4:00:01,  1.82s/it]
 46%|████▋     | 6827/14725 [3:29:50<3:57:51,  1.81s/it]
 46%|████▋     | 6828/14725 [3:29:52<3:57:52,  1.81s/it]
 46%|████▋     | 6829/14725 [3:29:53<3:56:24,  1.80s/it]
 46%|████▋     | 6830/14725 [3:29:55<3:55:21,  1.79s/it]
                                                        

 46%|████▋     | 6830/14725 [3:29:55
+0: {'loss': 0.8212, 'grad_norm': 0.728371412251643, 'learning_rate': 2e-05, 'epoch': 0.46}
+0: <3:55:21,  1.79s/it]
 46%|████▋     | 6831/14725 [3:29:57<3:56:14,  1.80s/it]
 46%|████▋     | 6832/14725 [3:29:59<3:56:08,  1.80s/it]
 46%|████▋     | 6833/14725 [3:30:01<4:01:46,  1.84s/it]
 46%|████▋     | 6834/14725 [3:30:03<4:00:45,  1.83s/it]
 46%|████▋     | 6835/14725 [3:30:04<4:01:37,  1.84s/it]
 46%|████▋     | 6836/14725 [3:30:06<4:02:00,  1.84s/it]
 46%|████▋     | 6837/14725 [3:30:08<3:59:43,  1.82s/it]
 46%|████▋     | 6838/14725 [3:30:10<3:57:35,  1.81s/it]
 46%|████▋     | 6839/14725 [3:30:12<3:56:00,  1.80s/it]
 46%|████▋     | 6840/14725 [3:30:13<3:55:06,  1.79s/it]
                                                        

 46%|████▋     | 6840/14725 [3:30:13<3:55:06,  1.79s/it]
 46%|████▋     | 6841/14725 [3:30:15<3:58:19,  1.81s/it]
 46%|████▋     | 6842/14725 [3:30:17<3:59:54,  1.83s/it]
 46%|████▋     | 6843/14725 [3:30:19<4:01:24,  1.84s/it]
 46%|�
+0: {'loss': 0.8206, 'grad_norm': 0.6966461105194368, 'learning_rate': 2e-05, 'epoch': 0.47}
+0: �███▋     | 6844/14725 [3:30:21<4:02:01,  1.84s/it]
 46%|████▋     | 6845/14725 [3:30:23<3:59:44,  1.83s/it]
 46%|████▋     | 6846/14725 [3:30:24<3:58:11,  1.81s/it]
 46%|████▋     | 6847/14725 [3:30:26<3:56:45,  1.80s/it]
 47%|████▋     | 6848/14725 [3:30:28<3:57:17,  1.81s/it]
 47%|████▋     | 6849/14725 [3:30:30<3:58:57,  1.82s/it]
 47%|████▋     | 6850/14725 [3:30:32<3:57:16,  1.81s/it]
                                                        

 47%|████▋     | 6850/14725 [3:30:32<3:57:16,  1.81s/it]
 47%|████▋     | 6851/14725 [3:30:33<3:56:49,  1.80s/it]
 47%|████▋     | 6852/14725 [3:30:35<3:57:19,  1.81s/it]
 47%|████▋     | 6853/14725 [3:30:37<3:58:43,  1.82s/it]
 47%|████▋     | 6854/14725 [3:30:39<3:56:18,  1.80s/it]
 47%|████▋     | 6855/14725 [3:30:41<3:56:33,  1.80s/it]
 47%|████▋     | 6856/14725 [3:30:43<4:02:50,  1.85s/it]
 47%|████▋     | 6857/147
+0: {'loss': 0.8233, 'grad_norm': 0.7094980504725136, 'learning_rate': 2e-05, 'epoch': 0.47}
+0: 25 [3:30:44<4:00:13,  1.83s/it]
 47%|████▋     | 6858/14725 [3:30:46<4:00:01,  1.83s/it]
 47%|████▋     | 6859/14725 [3:30:48<3:58:12,  1.82s/it]
 47%|████▋     | 6860/14725 [3:30:50<3:56:46,  1.81s/it]
                                                        

 47%|████▋     | 6860/14725 [3:30:50<3:56:46,  1.81s/it]
 47%|████▋     | 6861/14725 [3:30:52<3:56:20,  1.80s/it]
 47%|████▋     | 6862/14725 [3:30:53<3:56:43,  1.81s/it]
 47%|████▋     | 6863/14725 [3:30:55<3:55:13,  1.80s/it]
 47%|████▋     | 6864/14725 [3:30:57<3:55:30,  1.80s/it]
 47%|████▋     | 6865/14725 [3:30:59<3:56:09,  1.80s/it]
 47%|████▋     | 6866/14725 [3:31:01<3:54:32,  1.79s/it]
 47%|████▋     | 6867/14725 [3:31:02<3:54:46,  1.79s/it]
 47%|████▋     | 6868/14725 [3:31:04<3:52:40,  1.78s/it]
 47%|████▋     | 6869/14725 [3:31:06<3:53:19,  1.78s/it]
 47%|████▋     | 6870/14725 [3:31:08<3:56:43,  1.81s/
+0: {'loss': 0.814, 'grad_norm': 0.7235063436425769, 'learning_rate': 2e-05, 'epoch': 0.47}
+0: {'loss': 0.829, 'grad_norm': 0.6794602183371335, 'learning_rate': 2e-05, 'epoch': 0.47}
+0: it]
                                                        

 47%|████▋     | 6870/14725 [3:31:08<3:56:43,  1.81s/it]
 47%|████▋     | 6871/14725 [3:31:10<4:00:01,  1.83s/it]
 47%|████▋     | 6872/14725 [3:31:11<3:56:53,  1.81s/it]
 47%|████▋     | 6873/14725 [3:31:13<3:58:44,  1.82s/it]
 47%|████▋     | 6874/14725 [3:31:15<3:56:27,  1.81s/it]
 47%|████▋     | 6875/14725 [3:31:17<3:55:32,  1.80s/it]
 47%|████▋     | 6876/14725 [3:31:19<3:54:50,  1.80s/it]
 47%|████▋     | 6877/14725 [3:31:20<3:55:50,  1.80s/it]
 47%|████▋     | 6878/14725 [3:31:22<4:01:19,  1.85s/it]
 47%|████▋     | 6879/14725 [3:31:24<3:57:38,  1.82s/it]
 47%|████▋     | 6880/14725 [3:31:26<3:55:57,  1.80s/it]
                                                        

 47%|████▋     | 6880/14725 [3:31:26<3:55:57,  1.80s/it]
 47%|████▋     | 6881/14725 [3:31:28<3:58:05,  1.82s/it]
 47%|████▋     | 6882/1
+0: {'loss': 0.821, 'grad_norm': 0.7332500423314527, 'learning_rate': 2e-05, 'epoch': 0.47}
+0: 4725 [3:31:30<3:57:27,  1.82s/it]
 47%|████▋     | 6883/14725 [3:31:31<3:57:54,  1.82s/it]
 47%|████▋     | 6884/14725 [3:31:33<4:02:58,  1.86s/it]
 47%|████▋     | 6885/14725 [3:31:35<4:01:37,  1.85s/it]
 47%|████▋     | 6886/14725 [3:31:37<3:58:15,  1.82s/it]
 47%|████▋     | 6887/14725 [3:31:39<3:59:26,  1.83s/it]
 47%|████▋     | 6888/14725 [3:31:41<3:57:42,  1.82s/it]
 47%|████▋     | 6889/14725 [3:31:42<3:56:54,  1.81s/it]
 47%|████▋     | 6890/14725 [3:31:44<4:04:19,  1.87s/it]
                                                        

 47%|████▋     | 6890/14725 [3:31:44<4:04:19,  1.87s/it]
 47%|████▋     | 6891/14725 [3:31:46<4:01:50,  1.85s/it]
 47%|████▋     | 6892/14725 [3:31:48<3:59:14,  1.83s/it]
 47%|████▋     | 6893/14725 [3:31:50<3:58:08,  1.82s/it]
 47%|████▋     | 6894/14725 [3:31:52<3:55:16,  1.80s/it]
 47%|████▋     | 6895/14725 [3:31:53<3:55:08,  1.80
+0: {'loss': 0.8277, 'grad_norm': 0.7431408685126294, 'learning_rate': 2e-05, 'epoch': 0.47}
+0: s/it]
 47%|████▋     | 6896/14725 [3:31:55<3:54:00,  1.79s/it]
 47%|████▋     | 6897/14725 [3:31:57<3:55:28,  1.80s/it]
 47%|████▋     | 6898/14725 [3:31:59<3:59:25,  1.84s/it]
 47%|████▋     | 6899/14725 [3:32:01<3:58:35,  1.83s/it]
 47%|████▋     | 6900/14725 [3:32:03<4:16:24,  1.97s/it]
                                                        

 47%|████▋     | 6900/14725 [3:32:03<4:16:24,  1.97s/it]
 47%|████▋     | 6901/14725 [3:32:05<4:09:59,  1.92s/it]
 47%|████▋     | 6902/14725 [3:32:07<4:06:19,  1.89s/it]
 47%|████▋     | 6903/14725 [3:32:08<4:01:41,  1.85s/it]
 47%|████▋     | 6904/14725 [3:32:10<3:58:42,  1.83s/it]
 47%|████▋     | 6905/14725 [3:32:12<3:56:00,  1.81s/it]
 47%|████▋     | 6906/14725 [3:32:14<3:53:41,  1.79s/it]
 47%|████▋     | 6907/14725 [3:32:15<3:54:15,  1.80s/it]
 47%|████▋     | 6908/14725 [3:32:17<3:54:21,  1.80s/it]
 47%|████▋  
+0: {'loss': 0.825, 'grad_norm': 0.7682175098396979, 'learning_rate': 2e-05, 'epoch': 0.47}
+0: {'loss': 0.8091, 'grad_norm': 0.800993482356133, 'learning_rate': 2e-05, 'epoch': 0.47}
+0:    | 6909/14725 [3:32:19<3:54:50,  1.80s/it]
 47%|████▋     | 6910/14725 [3:32:21<3:54:55,  1.80s/it]
                                                        

 47%|████▋     | 6910/14725 [3:32:21<3:54:55,  1.80s/it]
 47%|████▋     | 6911/14725 [3:32:23<3:54:37,  1.80s/it]
 47%|████▋     | 6912/14725 [3:32:24<3:54:51,  1.80s/it]
 47%|████▋     | 6913/14725 [3:32:26<3:58:31,  1.83s/it]
 47%|████▋     | 6914/14725 [3:32:28<3:55:41,  1.81s/it]
 47%|████▋     | 6915/14725 [3:32:30<3:58:20,  1.83s/it]
 47%|████▋     | 6916/14725 [3:32:32<3:56:28,  1.82s/it]
 47%|████▋     | 6917/14725 [3:32:34<3:56:20,  1.82s/it]
 47%|████▋     | 6918/14725 [3:32:35<3:53:41,  1.80s/it]
 47%|████▋     | 6919/14725 [3:32:37<3:54:01,  1.80s/it]
 47%|████▋     | 6920/14725 [3:32:39<3:52:38,  1.79s/it]
                                                        

 47%|████▋     | 6920/14725 [3:32:39<3:52:38,  1.
+0: {'loss': 0.8117, 'grad_norm': 0.7945309215916032, 'learning_rate': 2e-05, 'epoch': 0.47}
+0: 79s/it]
 47%|████▋     | 6921/14725 [3:32:41<3:52:07,  1.78s/it]
 47%|████▋     | 6922/14725 [3:32:42<3:50:20,  1.77s/it]
 47%|████▋     | 6923/14725 [3:32:44<3:49:24,  1.76s/it]
 47%|████▋     | 6924/14725 [3:32:46<3:49:33,  1.77s/it]
 47%|████▋     | 6925/14725 [3:32:48<4:03:32,  1.87s/it]
 47%|████▋     | 6926/14725 [3:32:50<4:00:14,  1.85s/it]
 47%|████▋     | 6927/14725 [3:32:52<3:59:23,  1.84s/it]
 47%|████▋     | 6928/14725 [3:32:54<3:58:47,  1.84s/it]
 47%|████▋     | 6929/14725 [3:32:55<3:57:29,  1.83s/it]
 47%|████▋     | 6930/14725 [3:32:57<4:10:50,  1.93s/it]
                                                        

 47%|████▋     | 6930/14725 [3:32:57<4:10:50,  1.93s/it]
 47%|████▋     | 6931/14725 [3:32:59<4:07:00,  1.90s/it]
 47%|████▋     | 6932/14725 [3:33:01<4:04:17,  1.88s/it]
 47%|████▋     | 6933/14725 [3:33:03<4:01:09,  1.86s/it]
 47%|████▋
+0: {'loss': 0.8055, 'grad_norm': 0.7911246229247056, 'learning_rate': 2e-05, 'epoch': 0.47}
+0:      | 6934/14725 [3:33:05<3:58:23,  1.84s/it]
 47%|████▋     | 6935/14725 [3:33:07<3:56:34,  1.82s/it]
 47%|████▋     | 6936/14725 [3:33:08<3:55:52,  1.82s/it]
 47%|████▋     | 6937/14725 [3:33:10<3:55:52,  1.82s/it]
 47%|████▋     | 6938/14725 [3:33:12<3:54:46,  1.81s/it]
 47%|████▋     | 6939/14725 [3:33:14<3:52:53,  1.79s/it]
 47%|████▋     | 6940/14725 [3:33:16<3:53:28,  1.80s/it]
                                                        

 47%|████▋     | 6940/14725 [3:33:16<3:53:28,  1.80s/it]
 47%|████▋     | 6941/14725 [3:33:17<3:54:07,  1.80s/it]
 47%|████▋     | 6942/14725 [3:33:19<3:51:46,  1.79s/it]
 47%|████▋     | 6943/14725 [3:33:21<3:53:21,  1.80s/it]
 47%|████▋     | 6944/14725 [3:33:23<3:56:09,  1.82s/it]
 47%|████▋     | 6945/14725 [3:33:25<3:54:41,  1.81s/it]
 47%|████▋     | 6946/14725 [3:33:26<3:55:53,  1.82s/it]
 47%|████▋     | 6947/14725 [3:33:28<3
+0: {'loss': 0.8197, 'grad_norm': 0.7201668341675055, 'learning_rate': 2e-05, 'epoch': 0.47}
+0: :57:41,  1.83s/it]
 47%|████▋     | 6948/14725 [3:33:30<3:56:07,  1.82s/it]
 47%|████▋     | 6949/14725 [3:33:32<3:54:49,  1.81s/it]
 47%|████▋     | 6950/14725 [3:33:34<3:56:57,  1.83s/it]
                                                        

 47%|████▋     | 6950/14725 [3:33:34<3:56:57,  1.83s/it]
 47%|████▋     | 6951/14725 [3:33:36<3:55:48,  1.82s/it]
 47%|████▋     | 6952/14725 [3:33:37<3:54:24,  1.81s/it]
 47%|████▋     | 6953/14725 [3:33:39<3:53:08,  1.80s/it]
 47%|████▋     | 6954/14725 [3:33:41<3:53:23,  1.80s/it]
 47%|████▋     | 6955/14725 [3:33:43<4:09:06,  1.92s/it]
 47%|████▋     | 6956/14725 [3:33:45<4:04:32,  1.89s/it]
 47%|████▋     | 6957/14725 [3:33:47<3:58:49,  1.84s/it]
 47%|████▋     | 6958/14725 [3:33:48<3:55:59,  1.82s/it]
 47%|████▋     | 6959/14725 [3:33:50<3:53:58,  1.81s/it]
 47%|████▋     | 6960/14725 [3:33:52<3:56:37,  1.83s/it]
         
+0: {'loss': 0.806, 'grad_norm': 0.7096206062418303, 'learning_rate': 2e-05, 'epoch': 0.47}
+0: {'loss': 0.7914, 'grad_norm': 0.7066915099434902, 'learning_rate': 2e-05, 'epoch': 0.47}
+0:                                                

 47%|████▋     | 6960/14725 [3:33:52<3:56:37,  1.83s/it]
 47%|████▋     | 6961/14725 [3:33:54<3:55:26,  1.82s/it]
 47%|████▋     | 6962/14725 [3:33:56<3:53:03,  1.80s/it]
 47%|████▋     | 6963/14725 [3:33:57<3:52:50,  1.80s/it]
 47%|████▋     | 6964/14725 [3:33:59<3:53:53,  1.81s/it]
 47%|████▋     | 6965/14725 [3:34:01<3:54:53,  1.82s/it]
 47%|████▋     | 6966/14725 [3:34:03<3:53:57,  1.81s/it]
 47%|████▋     | 6967/14725 [3:34:05<3:53:30,  1.81s/it]
 47%|████▋     | 6968/14725 [3:34:06<3:51:25,  1.79s/it]
 47%|████▋     | 6969/14725 [3:34:08<3:51:45,  1.79s/it]
 47%|████▋     | 6970/14725 [3:34:10<3:52:22,  1.80s/it]
                                                        

 47%|████▋     | 6970/14725 [3:34:10<3:52:22,  1.80s/it]
 47%|████▋     | 6971/14725 [3:34:12<3:55:51,  1.83s/it]
 47%|████▋     | 6972/14725 [3:34:14
+0: {'loss': 0.816, 'grad_norm': 0.7451736864006037, 'learning_rate': 2e-05, 'epoch': 0.47}
+0: <3:54:24,  1.81s/it]
 47%|████▋     | 6973/14725 [3:34:15<3:52:06,  1.80s/it]
 47%|████▋     | 6974/14725 [3:34:17<3:51:19,  1.79s/it]
 47%|████▋     | 6975/14725 [3:34:19<3:52:35,  1.80s/it]
 47%|████▋     | 6976/14725 [3:34:21<3:52:11,  1.80s/it]
 47%|████▋     | 6977/14725 [3:34:23<3:52:49,  1.80s/it]
 47%|████▋     | 6978/14725 [3:34:24<3:51:11,  1.79s/it]
 47%|████▋     | 6979/14725 [3:34:26<3:49:48,  1.78s/it]
 47%|████▋     | 6980/14725 [3:34:28<3:50:01,  1.78s/it]
                                                        

 47%|████▋     | 6980/14725 [3:34:28<3:50:01,  1.78s/it]
 47%|████▋     | 6981/14725 [3:34:30<3:52:01,  1.80s/it]
 47%|████▋     | 6982/14725 [3:34:32<3:54:20,  1.82s/it]
 47%|████▋     | 6983/14725 [3:34:34<3:54:32,  1.82s/it]
 47%|████▋     | 6984/14725 [3:34:35<3:53:45,  1.81s/it]
 47%|████▋     | 6985/14725 [3:34:37<3:54:51,  1.82s/it]
 47%|�
+0: {'loss': 0.8185, 'grad_norm': 0.6978424884810597, 'learning_rate': 2e-05, 'epoch': 0.47}
+0: �███▋     | 6986/14725 [3:34:39<3:58:08,  1.85s/it]
 47%|████▋     | 6987/14725 [3:34:41<3:57:52,  1.84s/it]
 47%|████▋     | 6988/14725 [3:34:43<3:59:17,  1.86s/it]
 47%|████▋     | 6989/14725 [3:34:45<3:57:48,  1.84s/it]
 47%|████▋     | 6990/14725 [3:34:46<3:57:16,  1.84s/it]
                                                        

 47%|████▋     | 6990/14725 [3:34:46<3:57:16,  1.84s/it]
 47%|████▋     | 6991/14725 [3:34:48<3:54:55,  1.82s/it]
 47%|████▋     | 6992/14725 [3:34:50<3:55:08,  1.82s/it]
 47%|████▋     | 6993/14725 [3:34:52<3:54:23,  1.82s/it]
 47%|████▋     | 6994/14725 [3:34:54<3:52:22,  1.80s/it]
 48%|████▊     | 6995/14725 [3:34:55<3:50:24,  1.79s/it]
 48%|████▊     | 6996/14725 [3:34:57<3:51:39,  1.80s/it]
 48%|████▊     | 6997/14725 [3:34:59<3:53:50,  1.82s/it]
 48%|████▊     | 6998/14725 [3:35:01<3:54:09,  1.82s/it]
 48%|████▊     | 6999/147
+0: {'loss': 0.8264, 'grad_norm': 0.6834684738316129, 'learning_rate': 2e-05, 'epoch': 0.48}
+0: {'loss': 0.8389, 'grad_norm': 0.7770352281345994, 'learning_rate': 2e-05, 'epoch': 0.48}
+0: 25 [3:35:03<3:55:39,  1.83s/it]
 48%|████▊     | 7000/14725 [3:35:04<3:52:45,  1.81s/it]
                                                        

 48%|████▊     | 7000/14725 [3:35:04<3:52:45,  1.81s/it]
 48%|████▊     | 7001/14725 [3:35:06<3:53:14,  1.81s/it]
 48%|████▊     | 7002/14725 [3:35:08<3:53:35,  1.81s/it]
 48%|████▊     | 7003/14725 [3:35:10<3:54:38,  1.82s/it]
 48%|████▊     | 7004/14725 [3:35:12<3:54:10,  1.82s/it]
 48%|████▊     | 7005/14725 [3:35:14<3:55:36,  1.83s/it]
 48%|████▊     | 7006/14725 [3:35:15<3:56:44,  1.84s/it]
 48%|████▊     | 7007/14725 [3:35:17<3:54:04,  1.82s/it]
 48%|████▊     | 7008/14725 [3:35:19<3:52:27,  1.81s/it]
 48%|████▊     | 7009/14725 [3:35:21<3:53:14,  1.81s/it]
 48%|████▊     | 7010/14725 [3:35:23<3:53:00,  1.81s/it]
                                                        

 48%|████▊     | 7010/14725 [3:35:23<3:53:00,  1.81s/it]
 48%|
+0: {'loss': 0.8162, 'grad_norm': 0.7434594551831226, 'learning_rate': 2e-05, 'epoch': 0.48}
+0: ████▊     | 7011/14725 [3:35:24<3:53:21,  1.82s/it]
 48%|████▊     | 7012/14725 [3:35:26<3:51:38,  1.80s/it]
 48%|████▊     | 7013/14725 [3:35:28<3:52:08,  1.81s/it]
 48%|████▊     | 7014/14725 [3:35:30<3:49:34,  1.79s/it]
 48%|████▊     | 7015/14725 [3:35:32<3:49:26,  1.79s/it]
 48%|████▊     | 7016/14725 [3:35:33<3:50:39,  1.80s/it]
 48%|████▊     | 7017/14725 [3:35:35<3:48:55,  1.78s/it]
 48%|████▊     | 7018/14725 [3:35:37<3:52:00,  1.81s/it]
 48%|████▊     | 7019/14725 [3:35:39<3:49:56,  1.79s/it]
 48%|████▊     | 7020/14725 [3:35:41<3:50:57,  1.80s/it]
                                                        

 48%|████▊     | 7020/14725 [3:35:41<3:50:57,  1.80s/it]
 48%|████▊     | 7021/14725 [3:35:42<3:52:27,  1.81s/it]
 48%|████▊     | 7022/14725 [3:35:44<3:54:23,  1.83s/it]
 48%|████▊     | 7023/14725 [3:35:46<3:52:17,  1.81s/it]
 48%|████▊     | 7024/1
+0: {'loss': 0.8209, 'grad_norm': 0.744023452029774, 'learning_rate': 2e-05, 'epoch': 0.48}
+0: 4725 [3:35:48<3:53:25,  1.82s/it]
 48%|████▊     | 7025/14725 [3:35:50<3:52:53,  1.81s/it]
 48%|████▊     | 7026/14725 [3:35:52<3:55:07,  1.83s/it]
 48%|████▊     | 7027/14725 [3:35:54<3:57:54,  1.85s/it]
 48%|████▊     | 7028/14725 [3:35:55<3:54:17,  1.83s/it]
 48%|████▊     | 7029/14725 [3:35:57<3:51:49,  1.81s/it]
 48%|████▊     | 7030/14725 [3:35:59<3:54:31,  1.83s/it]
                                                        

 48%|████▊     | 7030/14725 [3:35:59<3:54:31,  1.83s/it]
 48%|████▊     | 7031/14725 [3:36:01<3:55:05,  1.83s/it]
 48%|████▊     | 7032/14725 [3:36:03<3:57:10,  1.85s/it]
 48%|████▊     | 7033/14725 [3:36:04<3:56:18,  1.84s/it]
 48%|████▊     | 7034/14725 [3:36:06<3:53:34,  1.82s/it]
 48%|████▊     | 7035/14725 [3:36:08<3:52:13,  1.81s/it]
 48%|████▊     | 7036/14725 [3:36:10<3:52:32,  1.81s/it]
 48%|████▊     | 7037/14725 [3:36:12<3:53:43,  1.82
+0: {'loss': 0.8164, 'grad_norm': 0.8019190726665545, 'learning_rate': 2e-05, 'epoch': 0.48}
+0: s/it]
 48%|████▊     | 7038/14725 [3:36:14<3:56:56,  1.85s/it]
 48%|████▊     | 7039/14725 [3:36:15<3:53:56,  1.83s/it]
 48%|████▊     | 7040/14725 [3:36:17<3:55:16,  1.84s/it]
                                                        

 48%|████▊     | 7040/14725 [3:36:17<3:55:16,  1.84s/it]
 48%|████▊     | 7041/14725 [3:36:19<3:53:30,  1.82s/it]
 48%|████▊     | 7042/14725 [3:36:21<3:50:28,  1.80s/it]
 48%|████▊     | 7043/14725 [3:36:23<3:50:16,  1.80s/it]
 48%|████▊     | 7044/14725 [3:36:24<3:49:59,  1.80s/it]
 48%|████▊     | 7045/14725 [3:36:26<3:53:10,  1.82s/it]
 48%|████▊     | 7046/14725 [3:36:28<3:54:05,  1.83s/it]
 48%|████▊     | 7047/14725 [3:36:30<3:53:33,  1.83s/it]
 48%|████▊     | 7048/14725 [3:36:32<3:55:55,  1.84s/it]
 48%|████▊     | 7049/14725 [3:36:34<3:53:20,  1.82s/it]
 48%|████▊     | 7050/14725 [3:36:35<3:52:36,  1.82s/it]
                      
+0: {'loss': 0.8477, 'grad_norm': 0.7201403941216449, 'learning_rate': 2e-05, 'epoch': 0.48}
+0: {'loss': 0.8748, 'grad_norm': 0.7599060184806915, 'learning_rate': 2e-05, 'epoch': 0.48}
+0:                                   

 48%|████▊     | 7050/14725 [3:36:35<3:52:36,  1.82s/it]
 48%|████▊     | 7051/14725 [3:36:37<3:54:23,  1.83s/it]
 48%|████▊     | 7052/14725 [3:36:39<3:53:14,  1.82s/it]
 48%|████▊     | 7053/14725 [3:36:41<3:53:51,  1.83s/it]
 48%|████▊     | 7054/14725 [3:36:43<3:51:47,  1.81s/it]
 48%|████▊     | 7055/14725 [3:36:44<3:50:11,  1.80s/it]
 48%|████▊     | 7056/14725 [3:36:46<3:49:38,  1.80s/it]
 48%|████▊     | 7057/14725 [3:36:48<3:49:56,  1.80s/it]
 48%|████▊     | 7058/14725 [3:36:50<3:51:37,  1.81s/it]
 48%|████▊     | 7059/14725 [3:36:52<3:50:46,  1.81s/it]
 48%|████▊     | 7060/14725 [3:36:53<3:50:12,  1.80s/it]
                                                        

 48%|████▊     | 7060/14725 [3:36:53<3:50:12,  1.80s/it]
 48%|████▊     | 7061/14725 [3:36:55<3:50:13,  1.80s/it]
 48%|████▊     | 7062/14725 [3:36:57<3:48:24,  1.
+0: {'loss': 0.824, 'grad_norm': 0.7021506718795809, 'learning_rate': 2e-05, 'epoch': 0.48}
+0: 79s/it]
 48%|████▊     | 7063/14725 [3:36:59<3:48:28,  1.79s/it]
 48%|████▊     | 7064/14725 [3:37:01<3:48:11,  1.79s/it]
 48%|████▊     | 7065/14725 [3:37:02<3:48:53,  1.79s/it]
 48%|████▊     | 7066/14725 [3:37:04<3:48:13,  1.79s/it]
 48%|████▊     | 7067/14725 [3:37:06<3:50:37,  1.81s/it]
 48%|████▊     | 7068/14725 [3:37:08<3:55:01,  1.84s/it]
 48%|████▊     | 7069/14725 [3:37:10<3:56:31,  1.85s/it]
 48%|████▊     | 7070/14725 [3:37:12<3:53:04,  1.83s/it]
                                                        

 48%|████▊     | 7070/14725 [3:37:12<3:53:04,  1.83s/it]
 48%|████▊     | 7071/14725 [3:37:13<3:54:49,  1.84s/it]
 48%|████▊     | 7072/14725 [3:37:15<3:52:40,  1.82s/it]
 48%|████▊     | 7073/14725 [3:37:17<3:52:36,  1.82s/it]
 48%|████▊     | 7074/14725 [3:37:19<3:53:29,  1.83s/it]
 48%|████▊     | 7075/14725 [3:37:21<3:54:35,  1.84s/it]
 48%|████▊
+0: {'loss': 0.8327, 'grad_norm': 0.7422413076096266, 'learning_rate': 2e-05, 'epoch': 0.48}
+0:      | 7076/14725 [3:37:23<3:51:39,  1.82s/it]
 48%|████▊     | 7077/14725 [3:37:24<3:49:51,  1.80s/it]
 48%|████▊     | 7078/14725 [3:37:26<3:52:01,  1.82s/it]
 48%|████▊     | 7079/14725 [3:37:28<3:52:51,  1.83s/it]
 48%|████▊     | 7080/14725 [3:37:30<3:50:17,  1.81s/it]
                                                        

 48%|████▊     | 7080/14725 [3:37:30<3:50:17,  1.81s/it]
 48%|████▊     | 7081/14725 [3:37:32<3:51:57,  1.82s/it]
 48%|████▊     | 7082/14725 [3:37:33<3:52:06,  1.82s/it]
 48%|████▊     | 7083/14725 [3:37:35<3:53:03,  1.83s/it]
 48%|████▊     | 7084/14725 [3:37:37<3:50:36,  1.81s/it]
 48%|████▊     | 7085/14725 [3:37:39<3:48:08,  1.79s/it]
 48%|████▊     | 7086/14725 [3:37:41<3:49:53,  1.81s/it]
 48%|████▊     | 7087/14725 [3:37:42<3:47:08,  1.78s/it]
 48%|████▊     | 7088/14725 [3:37:44<3:46:26,  1.78s/it]
 48%|████▊     | 7089/14725 [3:37:46<3
+0: {'loss': 0.8369, 'grad_norm': 0.722902104531266, 'learning_rate': 2e-05, 'epoch': 0.48}
+0: {'loss': 0.8298, 'grad_norm': 0.8077982378751809, 'learning_rate': 2e-05, 'epoch': 0.48}
+0: :51:18,  1.82s/it]
 48%|████▊     | 7090/14725 [3:37:48<3:52:13,  1.83s/it]
                                                        

 48%|████▊     | 7090/14725 [3:37:48<3:52:13,  1.83s/it]
 48%|████▊     | 7091/14725 [3:37:50<3:56:08,  1.86s/it]
 48%|████▊     | 7092/14725 [3:37:52<3:54:22,  1.84s/it]
 48%|████▊     | 7093/14725 [3:37:53<3:50:53,  1.82s/it]
 48%|████▊     | 7094/14725 [3:37:55<3:52:03,  1.82s/it]
 48%|████▊     | 7095/14725 [3:37:57<3:50:01,  1.81s/it]
 48%|████▊     | 7096/14725 [3:37:59<3:52:05,  1.83s/it]
 48%|████▊     | 7097/14725 [3:38:01<3:50:26,  1.81s/it]
 48%|████▊     | 7098/14725 [3:38:03<3:53:56,  1.84s/it]
 48%|████▊     | 7099/14725 [3:38:04<3:52:05,  1.83s/it]
 48%|████▊     | 7100/14725 [3:38:06<4:00:52,  1.90s/it]
                                                        

 48%|████▊     | 7100/14725 [3:38:06<4:00:52,  1.90s/it]
 48%|████�
+0: {'loss': 0.8443, 'grad_norm': 0.7848632365786029, 'learning_rate': 2e-05, 'epoch': 0.48}
+0: ��     | 7101/14725 [3:38:08<3:57:01,  1.87s/it]
 48%|████▊     | 7102/14725 [3:38:10<3:53:24,  1.84s/it]
 48%|████▊     | 7103/14725 [3:38:12<3:51:13,  1.82s/it]
 48%|████▊     | 7104/14725 [3:38:14<3:49:06,  1.80s/it]
 48%|████▊     | 7105/14725 [3:38:15<3:52:51,  1.83s/it]
 48%|████▊     | 7106/14725 [3:38:17<3:54:03,  1.84s/it]
 48%|████▊     | 7107/14725 [3:38:19<3:51:51,  1.83s/it]
 48%|████▊     | 7108/14725 [3:38:21<3:51:22,  1.82s/it]
 48%|████▊     | 7109/14725 [3:38:23<3:52:58,  1.84s/it]
 48%|████▊     | 7110/14725 [3:38:25<3:55:39,  1.86s/it]
                                                        

 48%|████▊     | 7110/14725 [3:38:25<3:55:39,  1.86s/it]
 48%|████▊     | 7111/14725 [3:38:26<3:54:15,  1.85s/it]
 48%|████▊     | 7112/14725 [3:38:28<3:53:34,  1.84s/it]
 48%|████▊     | 7113/14725 [3:38:30<3:50:38,  1.82s/it]
 48%|████▊     | 7114/14725 [3:38:32
+0: {'loss': 0.8362, 'grad_norm': 0.7388134424671343, 'learning_rate': 2e-05, 'epoch': 0.48}
+0: <3:49:04,  1.81s/it]
 48%|████▊     | 7115/14725 [3:38:34<3:49:31,  1.81s/it]
 48%|████▊     | 7116/14725 [3:38:35<3:47:38,  1.80s/it]
 48%|████▊     | 7117/14725 [3:38:37<3:46:36,  1.79s/it]
 48%|████▊     | 7118/14725 [3:38:39<3:47:11,  1.79s/it]
 48%|████▊     | 7119/14725 [3:38:41<3:45:28,  1.78s/it]
 48%|████▊     | 7120/14725 [3:38:43<3:46:50,  1.79s/it]
                                                        

 48%|████▊     | 7120/14725 [3:38:43<3:46:50,  1.79s/it]
 48%|████▊     | 7121/14725 [3:38:44<3:46:37,  1.79s/it]
 48%|████▊     | 7122/14725 [3:38:46<3:48:40,  1.80s/it]
 48%|████▊     | 7123/14725 [3:38:48<3:47:42,  1.80s/it]
 48%|████▊     | 7124/14725 [3:38:50<3:49:16,  1.81s/it]
 48%|████▊     | 7125/14725 [3:38:52<3:48:35,  1.80s/it]
 48%|████▊     | 7126/14725 [3:38:53<3:47:51,  1.80s/it]
 48%|████▊     | 7127/14725 [3:38:55<3:48:49,  1.81s/it]
 48%|�
+0: {'loss': 0.8297, 'grad_norm': 0.6756200260551196, 'learning_rate': 2e-05, 'epoch': 0.48}
+0: �███▊     | 7128/14725 [3:38:57<3:48:18,  1.80s/it]
 48%|████▊     | 7129/14725 [3:38:59<3:47:33,  1.80s/it]
 48%|████▊     | 7130/14725 [3:39:01<3:46:59,  1.79s/it]
                                                        

 48%|████▊     | 7130/14725 [3:39:01<3:46:59,  1.79s/it]
 48%|████▊     | 7131/14725 [3:39:02<3:48:20,  1.80s/it]
 48%|████▊     | 7132/14725 [3:39:04<3:46:37,  1.79s/it]
 48%|████▊     | 7133/14725 [3:39:06<3:45:38,  1.78s/it]
 48%|████▊     | 7134/14725 [3:39:08<3:49:00,  1.81s/it]
 48%|████▊     | 7135/14725 [3:39:10<3:49:19,  1.81s/it]
 48%|████▊     | 7136/14725 [3:39:11<3:49:23,  1.81s/it]
 48%|████▊     | 7137/14725 [3:39:13<3:51:21,  1.83s/it]
 48%|████▊     | 7138/14725 [3:39:15<3:50:19,  1.82s/it]
 48%|████▊     | 7139/14725 [3:39:17<3:48:30,  1.81s/it]
 48%|████▊     | 7140/14725 [3:39:19<3:48:44,  1.81s/it]
                                   
+0: {'loss': 0.8154, 'grad_norm': 0.746168288826562, 'learning_rate': 2e-05, 'epoch': 0.48}
+0: {'loss': 0.8196, 'grad_norm': 0.6953091868510808, 'learning_rate': 2e-05, 'epoch': 0.49}
+0:                      

 48%|████▊     | 7140/14725 [3:39:19<3:48:44,  1.81s/it]
 48%|████▊     | 7141/14725 [3:39:21<3:53:12,  1.85s/it]
 49%|████▊     | 7142/14725 [3:39:22<3:51:50,  1.83s/it]
 49%|████▊     | 7143/14725 [3:39:24<3:52:26,  1.84s/it]
 49%|████▊     | 7144/14725 [3:39:26<3:53:42,  1.85s/it]
 49%|████▊     | 7145/14725 [3:39:28<3:50:41,  1.83s/it]
 49%|████▊     | 7146/14725 [3:39:30<3:48:44,  1.81s/it]
 49%|████▊     | 7147/14725 [3:39:32<3:50:55,  1.83s/it]
 49%|████▊     | 7148/14725 [3:39:33<3:51:51,  1.84s/it]
 49%|████▊     | 7149/14725 [3:39:35<3:53:01,  1.85s/it]
 49%|████▊     | 7150/14725 [3:39:37<3:52:54,  1.84s/it]
                                                        

 49%|████▊     | 7150/14725 [3:39:37<3:52:54,  1.84s/it]
 49%|████▊     | 7151/14725 [3:39:39<3:55:39,  1.87s/it]
 49%|████▊     | 7152/14725 [3:39:41<3:56:58,  1.88s/it]
 49%|
+0: {'loss': 0.8116, 'grad_norm': 0.7135429958434556, 'learning_rate': 2e-05, 'epoch': 0.49}
+0: ████▊     | 7153/14725 [3:39:43<3:52:19,  1.84s/it]
 49%|████▊     | 7154/14725 [3:39:45<3:52:34,  1.84s/it]
 49%|████▊     | 7155/14725 [3:39:46<3:51:15,  1.83s/it]
 49%|████▊     | 7156/14725 [3:39:48<3:52:11,  1.84s/it]
 49%|████▊     | 7157/14725 [3:39:50<3:51:41,  1.84s/it]
 49%|████▊     | 7158/14725 [3:39:52<3:49:45,  1.82s/it]
 49%|████▊     | 7159/14725 [3:39:54<3:47:09,  1.80s/it]
 49%|████▊     | 7160/14725 [3:39:55<3:46:32,  1.80s/it]
                                                        

 49%|████▊     | 7160/14725 [3:39:55<3:46:32,  1.80s/it]
 49%|████▊     | 7161/14725 [3:39:57<3:46:22,  1.80s/it]
 49%|████▊     | 7162/14725 [3:39:59<3:47:15,  1.80s/it]
 49%|████▊     | 7163/14725 [3:40:01<3:48:05,  1.81s/it]
 49%|████▊     | 7164/14725 [3:40:03<3:47:39,  1.81s/it]
 49%|████▊     | 7165/14725 [3:40:04<3:49:42,  1.82s/it]
 49%|████▊     | 7166/1
+0: {'loss': 0.8401, 'grad_norm': 0.7506211663836354, 'learning_rate': 2e-05, 'epoch': 0.49}
+0: 4725 [3:40:06<3:50:16,  1.83s/it]
 49%|████▊     | 7167/14725 [3:40:08<3:50:57,  1.83s/it]
 49%|████▊     | 7168/14725 [3:40:10<3:49:33,  1.82s/it]
 49%|████▊     | 7169/14725 [3:40:12<3:50:50,  1.83s/it]
 49%|████▊     | 7170/14725 [3:40:14<3:50:39,  1.83s/it]
                                                        

 49%|████▊     | 7170/14725 [3:40:14<3:50:39,  1.83s/it]
 49%|████▊     | 7171/14725 [3:40:16<3:54:45,  1.86s/it]
 49%|████▊     | 7172/14725 [3:40:17<3:53:28,  1.85s/it]
 49%|████▊     | 7173/14725 [3:40:19<3:51:41,  1.84s/it]
 49%|████▊     | 7174/14725 [3:40:21<3:51:28,  1.84s/it]
 49%|████▊     | 7175/14725 [3:40:23<3:49:58,  1.83s/it]
 49%|████▊     | 7176/14725 [3:40:25<3:47:27,  1.81s/it]
 49%|████▊     | 7177/14725 [3:40:26<3:47:13,  1.81s/it]
 49%|████▊     | 7178/14725 [3:40:28<3:49:43,  1.83s/it]
 49%|████▉     | 7179/14725 [3:40:30<3:51:18,  1.84
+0: {'loss': 0.8234, 'grad_norm': 0.7500698306516146, 'learning_rate': 2e-05, 'epoch': 0.49}
+0: {'loss': 0.8278, 'grad_norm': 0.7434386889365426, 'learning_rate': 2e-05, 'epoch': 0.49}
+0: s/it]
 49%|████▉     | 7180/14725 [3:40:32<3:56:28,  1.88s/it]
                                                        

 49%|████▉     | 7180/14725 [3:40:32<3:56:28,  1.88s/it]
 49%|████▉     | 7181/14725 [3:40:34<3:54:31,  1.87s/it]
 49%|████▉     | 7182/14725 [3:40:36<3:55:37,  1.87s/it]
 49%|████▉     | 7183/14725 [3:40:38<3:53:05,  1.85s/it]
 49%|████▉     | 7184/14725 [3:40:39<3:50:17,  1.83s/it]
 49%|████▉     | 7185/14725 [3:40:41<3:47:43,  1.81s/it]
 49%|████▉     | 7186/14725 [3:40:43<3:48:37,  1.82s/it]
 49%|████▉     | 7187/14725 [3:40:45<3:47:36,  1.81s/it]
 49%|████▉     | 7188/14725 [3:40:47<3:48:53,  1.82s/it]
 49%|████▉     | 7189/14725 [3:40:49<3:49:15,  1.83s/it]
 49%|████▉     | 7190/14725 [3:40:50<3:53:28,  1.86s/it]
                                                        

 49%|████▉     | 7190/14725 [3:40:51<3:53:28,  1.86s/it]
 49%|████▉     | 7191
+0: {'loss': 0.8094, 'grad_norm': 0.7245854416308785, 'learning_rate': 2e-05, 'epoch': 0.49}
+0: /14725 [3:40:52<3:51:57,  1.85s/it]
 49%|████▉     | 7192/14725 [3:40:54<3:48:32,  1.82s/it]
 49%|████▉     | 7193/14725 [3:40:56<3:48:31,  1.82s/it]
 49%|████▉     | 7194/14725 [3:40:58<3:47:46,  1.81s/it]
 49%|████▉     | 7195/14725 [3:41:00<3:48:27,  1.82s/it]
 49%|████▉     | 7196/14725 [3:41:01<3:47:00,  1.81s/it]
 49%|████▉     | 7197/14725 [3:41:03<3:46:24,  1.80s/it]
 49%|████▉     | 7198/14725 [3:41:05<3:46:02,  1.80s/it]
 49%|████▉     | 7199/14725 [3:41:07<3:45:20,  1.80s/it]
 49%|████▉     | 7200/14725 [3:41:08<3:45:54,  1.80s/it]
                                                        

 49%|████▉     | 7200/14725 [3:41:09<3:45:54,  1.80s/it]
 49%|████▉     | 7201/14725 [3:41:10<3:45:28,  1.80s/it]
 49%|████▉     | 7202/14725 [3:41:12<3:46:05,  1.80s/it]
 49%|████▉     | 7203/14725 [3:41:14<3:48:07,  1.82s/it]
 49%|████▉     | 7204/14725 [3:41:16<3:47:45,  1.
+0: {'loss': 0.8151, 'grad_norm': 0.7110572286988209, 'learning_rate': 2e-05, 'epoch': 0.49}
+0: 82s/it]
 49%|████▉     | 7205/14725 [3:41:18<3:45:19,  1.80s/it]
 49%|████▉     | 7206/14725 [3:41:19<3:46:11,  1.80s/it]
 49%|████▉     | 7207/14725 [3:41:21<3:44:29,  1.79s/it]
 49%|████▉     | 7208/14725 [3:41:23<3:44:39,  1.79s/it]
 49%|████▉     | 7209/14725 [3:41:25<3:44:49,  1.79s/it]
 49%|████▉     | 7210/14725 [3:41:27<3:47:09,  1.81s/it]
                                                        

 49%|████▉     | 7210/14725 [3:41:27<3:47:09,  1.81s/it]
 49%|████▉     | 7211/14725 [3:41:28<3:46:23,  1.81s/it]
 49%|████▉     | 7212/14725 [3:41:30<3:47:31,  1.82s/it]
 49%|████▉     | 7213/14725 [3:41:32<3:48:19,  1.82s/it]
 49%|████▉     | 7214/14725 [3:41:34<3:49:16,  1.83s/it]
 49%|████▉     | 7215/14725 [3:41:36<3:48:59,  1.83s/it]
 49%|████▉     | 7216/14725 [3:41:37<3:46:09,  1.81s/it]
 49%|████▉     | 7217/14725 [3:41:39<3:45:04,  1.80s/it]
 49%|████▉
+0: {'loss': 0.8362, 'grad_norm': 0.8085645841585882, 'learning_rate': 2e-05, 'epoch': 0.49}
+0:      | 7218/14725 [3:41:41<3:46:53,  1.81s/it]
 49%|████▉     | 7219/14725 [3:41:43<3:46:16,  1.81s/it]
 49%|████▉     | 7220/14725 [3:41:45<3:47:01,  1.81s/it]
                                                        

 49%|████▉     | 7220/14725 [3:41:45<3:47:01,  1.81s/it]
 49%|████▉     | 7221/14725 [3:41:47<3:48:09,  1.82s/it]
 49%|████▉     | 7222/14725 [3:41:48<3:47:53,  1.82s/it]
 49%|████▉     | 7223/14725 [3:41:50<3:46:09,  1.81s/it]
 49%|████▉     | 7224/14725 [3:41:52<3:48:49,  1.83s/it]
 49%|████▉     | 7225/14725 [3:41:54<3:49:13,  1.83s/it]
 49%|████▉     | 7226/14725 [3:41:56<3:46:52,  1.82s/it]
 49%|████▉     | 7227/14725 [3:41:57<3:44:49,  1.80s/it]
 49%|████▉     | 7228/14725 [3:41:59<3:46:37,  1.81s/it]
 49%|████▉     | 7229/14725 [3:42:01<3:47:59,  1.82s/it]
 49%|████▉     | 7230/14725 [3:42:03<3:50:38,  1.85s/it]
                                                
+0: {'loss': 0.8284, 'grad_norm': 0.7280266278034522, 'learning_rate': 2e-05, 'epoch': 0.49}
+0: {'loss': 0.8133, 'grad_norm': 0.7757250202571043, 'learning_rate': 2e-05, 'epoch': 0.49}
+0:         

 49%|████▉     | 7230/14725 [3:42:03<3:50:38,  1.85s/it]
 49%|████▉     | 7231/14725 [3:42:05<3:47:24,  1.82s/it]
 49%|████▉     | 7232/14725 [3:42:07<3:45:22,  1.80s/it]
 49%|████▉     | 7233/14725 [3:42:08<3:44:50,  1.80s/it]
 49%|████▉     | 7234/14725 [3:42:10<3:44:06,  1.80s/it]
 49%|████▉     | 7235/14725 [3:42:12<3:45:33,  1.81s/it]
 49%|████▉     | 7236/14725 [3:42:14<3:42:38,  1.78s/it]
 49%|████▉     | 7237/14725 [3:42:15<3:41:42,  1.78s/it]
 49%|████▉     | 7238/14725 [3:42:17<3:43:36,  1.79s/it]
 49%|████▉     | 7239/14725 [3:42:19<3:45:35,  1.81s/it]
 49%|████▉     | 7240/14725 [3:42:21<3:47:48,  1.83s/it]
                                                        

 49%|████▉     | 7240/14725 [3:42:21<3:47:48,  1.83s/it]
 49%|████▉     | 7241/14725 [3:42:23<3:47:17,  1.82s/it]
 49%|████▉     | 7242/14725 [3:42:25<3:46:55,  1.82s/it]
 49%|████�
+0: {'loss': 0.8285, 'grad_norm': 0.6736735458591139, 'learning_rate': 2e-05, 'epoch': 0.49}
+0: ��     | 7243/14725 [3:42:26<3:44:25,  1.80s/it]
 49%|████▉     | 7244/14725 [3:42:28<3:45:33,  1.81s/it]
 49%|████▉     | 7245/14725 [3:42:30<3:45:38,  1.81s/it]
 49%|████▉     | 7246/14725 [3:42:32<3:46:19,  1.82s/it]
 49%|████▉     | 7247/14725 [3:42:34<3:44:52,  1.80s/it]
 49%|████▉     | 7248/14725 [3:42:35<3:42:32,  1.79s/it]
 49%|████▉     | 7249/14725 [3:42:37<3:43:16,  1.79s/it]
 49%|████▉     | 7250/14725 [3:42:39<3:44:05,  1.80s/it]
                                                        

 49%|████▉     | 7250/14725 [3:42:39<3:44:05,  1.80s/it]
 49%|████▉     | 7251/14725 [3:42:41<3:48:02,  1.83s/it]
 49%|████▉     | 7252/14725 [3:42:43<3:45:05,  1.81s/it]
 49%|████▉     | 7253/14725 [3:42:44<3:43:22,  1.79s/it]
 49%|████▉     | 7254/14725 [3:42:46<3:42:21,  1.79s/it]
 49%|████▉     | 7255/14725 [3:42:48<3:45:06,  1.81s/it]
 49%|████▉     | 7256/14725 [3:42:50
+0: {'loss': 0.8302, 'grad_norm': 0.7780379072463545, 'learning_rate': 2e-05, 'epoch': 0.49}
+0: <3:44:38,  1.80s/it]
 49%|████▉     | 7257/14725 [3:42:52<3:44:07,  1.80s/it]
 49%|████▉     | 7258/14725 [3:42:53<3:43:34,  1.80s/it]
 49%|████▉     | 7259/14725 [3:42:55<3:44:06,  1.80s/it]
 49%|████▉     | 7260/14725 [3:42:57<3:43:44,  1.80s/it]
                                                        

 49%|████▉     | 7260/14725 [3:42:57<3:43:44,  1.80s/it]
 49%|████▉     | 7261/14725 [3:42:59<3:46:12,  1.82s/it]
 49%|████▉     | 7262/14725 [3:43:01<3:44:18,  1.80s/it]
 49%|████▉     | 7263/14725 [3:43:02<3:45:04,  1.81s/it]
 49%|████▉     | 7264/14725 [3:43:04<3:43:58,  1.80s/it]
 49%|████▉     | 7265/14725 [3:43:06<3:44:14,  1.80s/it]
 49%|████▉     | 7266/14725 [3:43:08<3:46:00,  1.82s/it]
 49%|████▉     | 7267/14725 [3:43:10<3:45:07,  1.81s/it]
 49%|████▉     | 7268/14725 [3:43:12<3:48:54,  1.84s/it]
 49%|████▉     | 7269/14725 [3:43:13<3:47:26,  1.83s/it]
 49%|�
+0: {'loss': 0.8049, 'grad_norm': 0.7513973290188027, 'learning_rate': 2e-05, 'epoch': 0.49}
+0: {'loss': 0.8122, 'grad_norm': 0.6779097905482466, 'learning_rate': 2e-05, 'epoch': 0.49}
+0: �███▉     | 7270/14725 [3:43:15<3:53:49,  1.88s/it]
                                                        

 49%|████▉     | 7270/14725 [3:43:15<3:53:49,  1.88s/it]
 49%|████▉     | 7271/14725 [3:43:17<3:50:25,  1.85s/it]
 49%|████▉     | 7272/14725 [3:43:19<3:50:45,  1.86s/it]
 49%|████▉     | 7273/14725 [3:43:21<3:49:52,  1.85s/it]
 49%|████▉     | 7274/14725 [3:43:23<3:47:49,  1.83s/it]
 49%|████▉     | 7275/14725 [3:43:25<3:48:16,  1.84s/it]
 49%|████▉     | 7276/14725 [3:43:26<3:47:38,  1.83s/it]
 49%|████▉     | 7277/14725 [3:43:28<3:55:10,  1.89s/it]
 49%|████▉     | 7278/14725 [3:43:30<3:52:10,  1.87s/it]
 49%|████▉     | 7279/14725 [3:43:32<3:49:23,  1.85s/it]
 49%|████▉     | 7280/14725 [3:43:34<3:49:45,  1.85s/it]
                                                        

 49%|████▉     | 7280/14725 [3:43:34<3:49:45,  1.85s/it]
 49%|████▉     | 7281/14725 [3:43:
+0: {'loss': 0.8069, 'grad_norm': 0.7371213818734131, 'learning_rate': 2e-05, 'epoch': 0.5}
+0: 36<3:48:37,  1.84s/it]
 49%|████▉     | 7282/14725 [3:43:38<3:47:47,  1.84s/it]
 49%|████▉     | 7283/14725 [3:43:39<3:46:20,  1.82s/it]
 49%|████▉     | 7284/14725 [3:43:41<3:46:08,  1.82s/it]
 49%|████▉     | 7285/14725 [3:43:43<3:44:45,  1.81s/it]
 49%|████▉     | 7286/14725 [3:43:45<3:43:39,  1.80s/it]
 49%|████▉     | 7287/14725 [3:43:47<3:45:55,  1.82s/it]
 49%|████▉     | 7288/14725 [3:43:48<3:45:39,  1.82s/it]
 50%|████▉     | 7289/14725 [3:43:50<3:43:59,  1.81s/it]
 50%|████▉     | 7290/14725 [3:43:52<3:42:05,  1.79s/it]
                                                        

 50%|████▉     | 7290/14725 [3:43:52<3:42:05,  1.79s/it]
 50%|████▉     | 7291/14725 [3:43:54<3:42:06,  1.79s/it]
 50%|████▉     | 7292/14725 [3:43:56<3:45:09,  1.82s/it]
 50%|████▉     | 7293/14725 [3:43:57<3:48:14,  1.84s/it]
 50%|████▉     | 7294/14725 [3:43:59<3:48:28,  1.84s/it]
 50%|
+0: {'loss': 0.8246, 'grad_norm': 0.7221461572901391, 'learning_rate': 2e-05, 'epoch': 0.5}
+0: ████▉     | 7295/14725 [3:44:01<3:44:44,  1.81s/it]
 50%|████▉     | 7296/14725 [3:44:03<3:43:36,  1.81s/it]
 50%|████▉     | 7297/14725 [3:44:05<3:42:44,  1.80s/it]
 50%|████▉     | 7298/14725 [3:44:06<3:41:06,  1.79s/it]
 50%|████▉     | 7299/14725 [3:44:08<3:41:44,  1.79s/it]
 50%|████▉     | 7300/14725 [3:44:10<3:40:48,  1.78s/it]
                                                        

 50%|████▉     | 7300/14725 [3:44:10<3:40:48,  1.78s/it]
 50%|████▉     | 7301/14725 [3:44:12<3:42:12,  1.80s/it]
 50%|████▉     | 7302/14725 [3:44:14<3:41:02,  1.79s/it]
 50%|████▉     | 7303/14725 [3:44:15<3:40:05,  1.78s/it]
 50%|████▉     | 7304/14725 [3:44:17<3:45:17,  1.82s/it]
 50%|████▉     | 7305/14725 [3:44:19<3:44:33,  1.82s/it]
 50%|████▉     | 7306/14725 [3:44:21<3:42:42,  1.80s/it]
 50%|████▉     | 7307/14725 [3:44:23<3:41:25,  1.79s/it]
 50%|████▉     | 7308/1
+0: {'loss': 0.8021, 'grad_norm': 0.7006250769471013, 'learning_rate': 2e-05, 'epoch': 0.5}
+0: {'loss': 0.8174, 'grad_norm': 0.7455410044859497, 'learning_rate': 2e-05, 'epoch': 0.5}
+0: 4725 [3:44:24<3:40:18,  1.78s/it]
 50%|████▉     | 7309/14725 [3:44:26<3:42:45,  1.80s/it]
 50%|████▉     | 7310/14725 [3:44:28<3:42:25,  1.80s/it]
                                                        

 50%|████▉     | 7310/14725 [3:44:28<3:42:25,  1.80s/it]
 50%|████▉     | 7311/14725 [3:44:30<3:44:31,  1.82s/it]
 50%|████▉     | 7312/14725 [3:44:32<3:42:01,  1.80s/it]
 50%|████▉     | 7313/14725 [3:44:33<3:42:20,  1.80s/it]
 50%|████▉     | 7314/14725 [3:44:35<3:43:43,  1.81s/it]
 50%|████▉     | 7315/14725 [3:44:37<3:45:46,  1.83s/it]
 50%|████▉     | 7316/14725 [3:44:39<3:42:51,  1.80s/it]
 50%|████▉     | 7317/14725 [3:44:41<3:44:21,  1.82s/it]
 50%|████▉     | 7318/14725 [3:44:42<3:43:22,  1.81s/it]
 50%|████▉     | 7319/14725 [3:44:44<3:41:54,  1.80s/it]
 50%|████▉     | 7320/14725 [3:44:46<3:43:23,  1.81s/it]
                                                        

 50
+0: {'loss': 0.8131, 'grad_norm': 0.7327242700131528, 'learning_rate': 2e-05, 'epoch': 0.5}
+0: %|████▉     | 7320/14725 [3:44:46<3:43:23,  1.81s/it]
 50%|████▉     | 7321/14725 [3:44:48<3:44:41,  1.82s/it]
 50%|████▉     | 7322/14725 [3:44:50<3:43:32,  1.81s/it]
 50%|████▉     | 7323/14725 [3:44:52<3:41:51,  1.80s/it]
 50%|████▉     | 7324/14725 [3:44:53<3:41:14,  1.79s/it]
 50%|████▉     | 7325/14725 [3:44:55<3:39:41,  1.78s/it]
 50%|████▉     | 7326/14725 [3:44:57<3:40:08,  1.79s/it]
 50%|████▉     | 7327/14725 [3:44:59<3:40:01,  1.78s/it]
 50%|████▉     | 7328/14725 [3:45:00<3:41:07,  1.79s/it]
 50%|████▉     | 7329/14725 [3:45:02<3:47:23,  1.84s/it]
 50%|████▉     | 7330/14725 [3:45:04<3:47:26,  1.85s/it]
                                                        

 50%|████▉     | 7330/14725 [3:45:04<3:47:26,  1.85s/it]
 50%|████▉     | 7331/14725 [3:45:06<3:53:21,  1.89s/it]
 50%|████▉     | 7332/14725 [3:45:08<3:51:02,  1.88s/it]
 50%|████▉     | 7333
+0: {'loss': 0.8202, 'grad_norm': 0.791825758586706, 'learning_rate': 2e-05, 'epoch': 0.5}
+0: /14725 [3:45:10<3:48:50,  1.86s/it]
 50%|████▉     | 7334/14725 [3:45:12<3:46:38,  1.84s/it]
 50%|████▉     | 7335/14725 [3:45:14<3:47:33,  1.85s/it]
 50%|████▉     | 7336/14725 [3:45:15<3:45:11,  1.83s/it]
 50%|████▉     | 7337/14725 [3:45:17<3:48:47,  1.86s/it]
 50%|████▉     | 7338/14725 [3:45:19<3:46:42,  1.84s/it]
 50%|████▉     | 7339/14725 [3:45:21<3:44:22,  1.82s/it]
 50%|████▉     | 7340/14725 [3:45:23<3:46:01,  1.84s/it]
                                                        

 50%|████▉     | 7340/14725 [3:45:23<3:46:01,  1.84s/it]
 50%|████▉     | 7341/14725 [3:45:25<3:44:01,  1.82s/it]
 50%|████▉     | 7342/14725 [3:45:26<3:42:52,  1.81s/it]
 50%|████▉     | 7343/14725 [3:45:28<3:41:53,  1.80s/it]
 50%|████▉     | 7344/14725 [3:45:30<3:41:16,  1.80s/it]
 50%|████▉     | 7345/14725 [3:45:32<3:39:48,  1.79s/it]
 50%|████▉     | 7346/14725 [3:45:34<3:44:15,  1.
+0: {'loss': 0.812, 'grad_norm': 0.6966635463599664, 'learning_rate': 2e-05, 'epoch': 0.5}
+0: 82s/it]
 50%|████▉     | 7347/14725 [3:45:35<3:42:42,  1.81s/it]
 50%|████▉     | 7348/14725 [3:45:37<3:42:34,  1.81s/it]
 50%|████▉     | 7349/14725 [3:45:39<3:46:54,  1.85s/it]
 50%|████▉     | 7350/14725 [3:45:41<3:44:27,  1.83s/it]
                                                        

 50%|████▉     | 7350/14725 [3:45:41<3:44:27,  1.83s/it]
 50%|████▉     | 7351/14725 [3:45:43<3:43:49,  1.82s/it]
 50%|████▉     | 7352/14725 [3:45:44<3:42:05,  1.81s/it]
 50%|████▉     | 7353/14725 [3:45:46<3:39:55,  1.79s/it]
 50%|████▉     | 7354/14725 [3:45:48<3:38:12,  1.78s/it]
 50%|████▉     | 7355/14725 [3:45:50<3:37:08,  1.77s/it]
 50%|████▉     | 7356/14725 [3:45:51<3:36:38,  1.76s/it]
 50%|████▉     | 7357/14725 [3:45:53<3:39:00,  1.78s/it]
 50%|████▉     | 7358/14725 [3:45:55<3:40:56,  1.80s/it]
 50%|████▉     | 7359/14725 [3:45:57<3:40:41,  1.80s/it]
 50%|████▉
+0: {'loss': 0.8152, 'grad_norm': 0.7145567998136748, 'learning_rate': 2e-05, 'epoch': 0.5}
+0: {'loss': 0.8166, 'grad_norm': 0.7026435118946406, 'learning_rate': 2e-05, 'epoch': 0.5}
+0:      | 7360/14725 [3:45:59<3:42:06,  1.81s/it]
                                                        

 50%|████▉     | 7360/14725 [3:45:59<3:42:06,  1.81s/it]
 50%|████▉     | 7361/14725 [3:46:01<3:42:23,  1.81s/it]
 50%|████▉     | 7362/14725 [3:46:02<3:44:07,  1.83s/it]
 50%|█████     | 7363/14725 [3:46:04<3:44:17,  1.83s/it]
 50%|█████     | 7364/14725 [3:46:06<3:43:11,  1.82s/it]
 50%|█████     | 7365/14725 [3:46:08<3:41:41,  1.81s/it]
 50%|█████     | 7366/14725 [3:46:10<3:41:37,  1.81s/it]
 50%|█████     | 7367/14725 [3:46:11<3:42:19,  1.81s/it]
 50%|█████     | 7368/14725 [3:46:13<3:39:46,  1.79s/it]
 50%|█████     | 7369/14725 [3:46:15<3:43:07,  1.82s/it]
 50%|█████     | 7370/14725 [3:46:17<3:44:41,  1.83s/it]
                                                        

 50%|█████     | 7370/14725 [3:46:17<3:44:41,  1.83s/it]
 50%|█████     | 7371/14725 [3:46:19<3:42:06,  
+0: {'loss': 0.8348, 'grad_norm': 0.7348084316917504, 'learning_rate': 2e-05, 'epoch': 0.5}
+0: 1.81s/it]
 50%|█████     | 7372/14725 [3:46:21<3:44:34,  1.83s/it]
 50%|█████     | 7373/14725 [3:46:22<3:45:34,  1.84s/it]
 50%|█████     | 7374/14725 [3:46:24<3:46:19,  1.85s/it]
 50%|█████     | 7375/14725 [3:46:26<3:48:10,  1.86s/it]
 50%|█████     | 7376/14725 [3:46:28<3:46:40,  1.85s/it]
 50%|█████     | 7377/14725 [3:46:30<3:45:00,  1.84s/it]
 50%|█████     | 7378/14725 [3:46:32<3:46:51,  1.85s/it]
 50%|█████     | 7379/14725 [3:46:34<3:46:56,  1.85s/it]
 50%|█████     | 7380/14725 [3:46:35<3:44:47,  1.84s/it]
                                                        

 50%|█████     | 7380/14725 [3:46:35<3:44:47,  1.84s/it]
 50%|█████     | 7381/14725 [3:46:37<3:46:34,  1.85s/it]
 50%|█████     | 7382/14725 [3:46:39<3:45:35,  1.84s/it]
 50%|█████     | 7383/14725 [3:46:41<3:42:54,  1.82s/it]
 50%|█████     | 7384/14725 [3:46:43<3:41:12,  1.81s/it]
 50%|████�
+0: {'loss': 0.8369, 'grad_norm': 0.7283785216507142, 'learning_rate': 2e-05, 'epoch': 0.5}
+0: ��     | 7385/14725 [3:46:44<3:39:49,  1.80s/it]
 50%|█████     | 7386/14725 [3:46:46<3:38:57,  1.79s/it]
 50%|█████     | 7387/14725 [3:46:48<3:38:34,  1.79s/it]
 50%|█████     | 7388/14725 [3:46:50<3:39:59,  1.80s/it]
 50%|█████     | 7389/14725 [3:46:52<3:40:31,  1.80s/it]
 50%|█████     | 7390/14725 [3:46:53<3:40:54,  1.81s/it]
                                                        

 50%|█████     | 7390/14725 [3:46:53<3:40:54,  1.81s/it]
 50%|█████     | 7391/14725 [3:46:55<3:41:21,  1.81s/it]
 50%|█████     | 7392/14725 [3:46:57<3:39:32,  1.80s/it]
 50%|█████     | 7393/14725 [3:46:59<3:41:44,  1.81s/it]
 50%|█████     | 7394/14725 [3:47:01<3:41:49,  1.82s/it]
 50%|█████     | 7395/14725 [3:47:02<3:40:44,  1.81s/it]
 50%|█████     | 7396/14725 [3:47:04<3:41:17,  1.81s/it]
 50%|█████     | 7397/14725 [3:47:06<3:40:46,  1.81s/it]
 50%|█████     | 7398/14725 [3:47:08
+0: {'loss': 0.8269, 'grad_norm': 0.8038789161048628, 'learning_rate': 2e-05, 'epoch': 0.5}
+0: {'loss': 0.8186, 'grad_norm': 0.7551822975600762, 'learning_rate': 2e-05, 'epoch': 0.5}
+0: <3:39:38,  1.80s/it]
 50%|█████     | 7399/14725 [3:47:10<3:41:44,  1.82s/it]
 50%|█████     | 7400/14725 [3:47:12<3:43:06,  1.83s/it]
                                                        

 50%|█████     | 7400/14725 [3:47:12<3:43:06,  1.83s/it]
 50%|█████     | 7401/14725 [3:47:13<3:42:07,  1.82s/it]
 50%|█████     | 7402/14725 [3:47:15<3:41:20,  1.81s/it]
 50%|█████     | 7403/14725 [3:47:17<3:39:20,  1.80s/it]
 50%|█████     | 7404/14725 [3:47:19<3:40:08,  1.80s/it]
 50%|█████     | 7405/14725 [3:47:21<3:41:36,  1.82s/it]
 50%|█████     | 7406/14725 [3:47:22<3:40:37,  1.81s/it]
 50%|█████     | 7407/14725 [3:47:24<3:39:05,  1.80s/it]
 50%|█████     | 7408/14725 [3:47:26<3:40:53,  1.81s/it]
 50%|█████     | 7409/14725 [3:47:28<3:39:42,  1.80s/it]
 50%|█████     | 7410/14725 [3:47:30<3:38:25,  1.79s/it]
                                                        

 50%|███�
+0: {'loss': 0.8049, 'grad_norm': 0.7013507837704557, 'learning_rate': 2e-05, 'epoch': 0.5}
+0: �█     | 7410/14725 [3:47:30<3:38:25,  1.79s/it]
 50%|█████     | 7411/14725 [3:47:31<3:42:50,  1.83s/it]
 50%|█████     | 7412/14725 [3:47:33<3:40:35,  1.81s/it]
 50%|█████     | 7413/14725 [3:47:35<3:38:02,  1.79s/it]
 50%|█████     | 7414/14725 [3:47:37<3:36:04,  1.77s/it]
 50%|█████     | 7415/14725 [3:47:38<3:36:35,  1.78s/it]
 50%|█████     | 7416/14725 [3:47:40<3:36:08,  1.77s/it]
 50%|█████     | 7417/14725 [3:47:42<3:38:19,  1.79s/it]
 50%|█████     | 7418/14725 [3:47:44<3:37:43,  1.79s/it]
 50%|█████     | 7419/14725 [3:47:46<3:37:10,  1.78s/it]
 50%|█████     | 7420/14725 [3:47:47<3:38:48,  1.80s/it]
                                                        

 50%|█████     | 7420/14725 [3:47:47<3:38:48,  1.80s/it]
 50%|█████     | 7421/14725 [3:47:49<3:41:37,  1.82s/it]
 50%|█████     | 7422/14725 [3:47:51<3:43:27,  1.84s/it]
 50%|█████     | 7423/14725 [3:47:
+0: {'loss': 0.8163, 'grad_norm': 0.7324453425363009, 'learning_rate': 2e-05, 'epoch': 0.5}
+0: 53<3:43:46,  1.84s/it]
 50%|█████     | 7424/14725 [3:47:55<3:42:21,  1.83s/it]
 50%|█████     | 7425/14725 [3:47:57<3:41:06,  1.82s/it]
 50%|█████     | 7426/14725 [3:47:58<3:39:03,  1.80s/it]
 50%|█████     | 7427/14725 [3:48:00<3:38:16,  1.79s/it]
 50%|█████     | 7428/14725 [3:48:02<3:41:19,  1.82s/it]
 50%|█████     | 7429/14725 [3:48:04<3:41:15,  1.82s/it]
 50%|█████     | 7430/14725 [3:48:06<3:40:18,  1.81s/it]
                                                        

 50%|█████     | 7430/14725 [3:48:06<3:40:18,  1.81s/it]
 50%|█████     | 7431/14725 [3:48:08<3:42:50,  1.83s/it]
 50%|█████     | 7432/14725 [3:48:09<3:43:51,  1.84s/it]
 50%|█████     | 7433/14725 [3:48:11<3:42:02,  1.83s/it]
 50%|█████     | 7434/14725 [3:48:13<3:41:11,  1.82s/it]
 50%|█████     | 7435/14725 [3:48:15<3:43:24,  1.84s/it]
 50%|█████     | 7436/14725 [3:48:17<3:43:51,  1.84s/it]
 51%|
+0: {'loss': 0.8195, 'grad_norm': 0.7198873798267366, 'learning_rate': 2e-05, 'epoch': 0.51}
+0: █████     | 7437/14725 [3:48:19<3:41:52,  1.83s/it]
 51%|█████     | 7438/14725 [3:48:21<3:56:27,  1.95s/it]
 51%|█████     | 7439/14725 [3:48:23<3:53:00,  1.92s/it]
 51%|█████     | 7440/14725 [3:48:24<3:50:54,  1.90s/it]
                                                        

 51%|█████     | 7440/14725 [3:48:25<3:50:54,  1.90s/it]
 51%|█████     | 7441/14725 [3:48:26<3:50:20,  1.90s/it]
 51%|█████     | 7442/14725 [3:48:28<3:46:22,  1.86s/it]
 51%|█████     | 7443/14725 [3:48:30<3:45:41,  1.86s/it]
 51%|█████     | 7444/14725 [3:48:32<3:44:56,  1.85s/it]
 51%|█████     | 7445/14725 [3:48:34<3:46:33,  1.87s/it]
 51%|█████     | 7446/14725 [3:48:36<3:42:50,  1.84s/it]
 51%|█████     | 7447/14725 [3:48:37<3:40:23,  1.82s/it]
 51%|█████     | 7448/14725 [3:48:39<3:40:38,  1.82s/it]
 51%|█████     | 7449/14725 [3:48:41<3:41:20,  1.83s/it]
 51%|█████     | 7450/1
+0: {'loss': 0.8098, 'grad_norm': 0.6818693027038021, 'learning_rate': 2e-05, 'epoch': 0.51}
+0: {'loss': 0.8508, 'grad_norm': 0.7407117253317056, 'learning_rate': 2e-05, 'epoch': 0.51}
+0: 4725 [3:48:43<3:42:14,  1.83s/it]
                                                        

 51%|█████     | 7450/14725 [3:48:43<3:42:14,  1.83s/it]
 51%|█████     | 7451/14725 [3:48:45<3:46:33,  1.87s/it]
 51%|█████     | 7452/14725 [3:48:47<3:43:18,  1.84s/it]
 51%|█████     | 7453/14725 [3:48:48<3:41:44,  1.83s/it]
 51%|█████     | 7454/14725 [3:48:50<3:44:34,  1.85s/it]
 51%|█████     | 7455/14725 [3:48:52<3:41:14,  1.83s/it]
 51%|█████     | 7456/14725 [3:48:54<3:41:28,  1.83s/it]
 51%|█████     | 7457/14725 [3:48:56<3:40:40,  1.82s/it]
 51%|█████     | 7458/14725 [3:48:57<3:41:35,  1.83s/it]
 51%|█████     | 7459/14725 [3:48:59<3:39:24,  1.81s/it]
 51%|█████     | 7460/14725 [3:49:01<3:39:19,  1.81s/it]
                                                        

 51%|█████     | 7460/14725 [3:49:01<3:39:19,  1.81s/it]
 51%|█████     | 7461/14725 [3:49:03<3:40:39,  1.82s/it]
 51
+0: {'loss': 0.8153, 'grad_norm': 0.7215908478881512, 'learning_rate': 2e-05, 'epoch': 0.51}
+0: %|█████     | 7462/14725 [3:49:05<3:39:44,  1.82s/it]
 51%|█████     | 7463/14725 [3:49:07<3:40:15,  1.82s/it]
 51%|█████     | 7464/14725 [3:49:08<3:39:24,  1.81s/it]
 51%|█████     | 7465/14725 [3:49:10<3:38:39,  1.81s/it]
 51%|█████     | 7466/14725 [3:49:12<3:39:44,  1.82s/it]
 51%|█████     | 7467/14725 [3:49:14<3:38:26,  1.81s/it]
 51%|█████     | 7468/14725 [3:49:16<3:39:54,  1.82s/it]
 51%|█████     | 7469/14725 [3:49:17<3:39:53,  1.82s/it]
 51%|█████     | 7470/14725 [3:49:19<3:41:11,  1.83s/it]
                                                        

 51%|█████     | 7470/14725 [3:49:19<3:41:11,  1.83s/it]
 51%|█████     | 7471/14725 [3:49:21<3:40:34,  1.82s/it]
 51%|█████     | 7472/14725 [3:49:23<3:39:51,  1.82s/it]
 51%|█████     | 7473/14725 [3:49:25<3:38:22,  1.81s/it]
 51%|█████     | 7474/14725 [3:49:26<3:38:36,  1.81s/it]
 51%|█████     | 7475
+0: {'loss': 0.8221, 'grad_norm': 0.7283779197650083, 'learning_rate': 2e-05, 'epoch': 0.51}
+0: /14725 [3:49:28<3:39:28,  1.82s/it]
 51%|█████     | 7476/14725 [3:49:30<3:38:02,  1.80s/it]
 51%|█████     | 7477/14725 [3:49:32<3:42:26,  1.84s/it]
 51%|█████     | 7478/14725 [3:49:34<3:43:02,  1.85s/it]
 51%|█████     | 7479/14725 [3:49:36<3:40:29,  1.83s/it]
 51%|█████     | 7480/14725 [3:49:37<3:39:26,  1.82s/it]
                                                        

 51%|█████     | 7480/14725 [3:49:37<3:39:26,  1.82s/it]
 51%|█████     | 7481/14725 [3:49:39<3:39:55,  1.82s/it]
 51%|█████     | 7482/14725 [3:49:41<3:39:32,  1.82s/it]
 51%|█████     | 7483/14725 [3:49:43<3:38:21,  1.81s/it]
 51%|█████     | 7484/14725 [3:49:45<3:40:31,  1.83s/it]
 51%|█████     | 7485/14725 [3:49:47<3:37:56,  1.81s/it]
 51%|█████     | 7486/14725 [3:49:48<3:35:49,  1.79s/it]
 51%|█████     | 7487/14725 [3:49:50<3:37:08,  1.80s/it]
 51%|█████     | 7488/14725 [3:49:52<3:35:03,  1.
+0: {'loss': 0.8165, 'grad_norm': 0.7467404945498635, 'learning_rate': 2e-05, 'epoch': 0.51}
+0: {'loss': 0.8005, 'grad_norm': 0.8007259130990007, 'learning_rate': 2e-05, 'epoch': 0.51}
+0: 78s/it]
 51%|█████     | 7489/14725 [3:49:54<3:35:42,  1.79s/it]
 51%|█████     | 7490/14725 [3:49:55<3:35:27,  1.79s/it]
                                                        

 51%|█████     | 7490/14725 [3:49:55<3:35:27,  1.79s/it]
 51%|█████     | 7491/14725 [3:49:57<3:34:48,  1.78s/it]
 51%|█████     | 7492/14725 [3:49:59<3:34:05,  1.78s/it]
 51%|█████     | 7493/14725 [3:50:01<3:33:27,  1.77s/it]
 51%|█████     | 7494/14725 [3:50:02<3:33:29,  1.77s/it]
 51%|█████     | 7495/14725 [3:50:04<3:33:21,  1.77s/it]
 51%|█████     | 7496/14725 [3:50:06<3:34:54,  1.78s/it]
 51%|█████     | 7497/14725 [3:50:08<3:35:35,  1.79s/it]
 51%|█████     | 7498/14725 [3:50:10<3:35:57,  1.79s/it]
 51%|█████     | 7499/14725 [3:50:12<3:38:28,  1.81s/it]
 51%|█████     | 7500/14725 [3:50:13<3:38:14,  1.81s/it]
                                                        

 51%|█████     | 75
+0: {'loss': 0.8351, 'grad_norm': 0.7379265156241763, 'learning_rate': 2e-05, 'epoch': 0.51}
+0: 00/14725 [3:50:13<3:38:14,  1.81s/it]
 51%|█████     | 7501/14725 [3:50:15<3:37:14,  1.80s/it]
 51%|█████     | 7502/14725 [3:50:17<3:36:04,  1.79s/it]
 51%|█████     | 7503/14725 [3:50:19<3:36:00,  1.79s/it]
 51%|█████     | 7504/14725 [3:50:21<3:36:25,  1.80s/it]
 51%|█████     | 7505/14725 [3:50:22<3:38:37,  1.82s/it]
 51%|█████     | 7506/14725 [3:50:24<3:37:24,  1.81s/it]
 51%|█████     | 7507/14725 [3:50:26<3:36:38,  1.80s/it]
 51%|█████     | 7508/14725 [3:50:28<3:37:44,  1.81s/it]
 51%|█████     | 7509/14725 [3:50:30<3:36:36,  1.80s/it]
 51%|█████     | 7510/14725 [3:50:31<3:35:32,  1.79s/it]
                                                        

 51%|█████     | 7510/14725 [3:50:31<3:35:32,  1.79s/it]
 51%|█████     | 7511/14725 [3:50:33<3:35:20,  1.79s/it]
 51%|█████     | 7512/14725 [3:50:35<3:35:42,  1.79s/it]
 51%|█████     | 7513/14725 [3:50:37<3:36:50,  
+0: {'loss': 0.8498, 'grad_norm': 0.6789950387551114, 'learning_rate': 2e-05, 'epoch': 0.51}
+0: 1.80s/it]
 51%|█████     | 7514/14725 [3:50:39<3:38:01,  1.81s/it]
 51%|█████     | 7515/14725 [3:50:40<3:36:21,  1.80s/it]
 51%|█████     | 7516/14725 [3:50:42<3:38:19,  1.82s/it]
 51%|█████     | 7517/14725 [3:50:44<3:36:17,  1.80s/it]
 51%|█████     | 7518/14725 [3:50:46<3:36:55,  1.81s/it]
 51%|█████     | 7519/14725 [3:50:48<3:35:39,  1.80s/it]
 51%|█████     | 7520/14725 [3:50:49<3:33:42,  1.78s/it]
                                                        

 51%|█████     | 7520/14725 [3:50:49<3:33:42,  1.78s/it]
 51%|█████     | 7521/14725 [3:50:51<3:35:22,  1.79s/it]
 51%|█████     | 7522/14725 [3:50:53<3:33:23,  1.78s/it]
 51%|█████     | 7523/14725 [3:50:55<3:33:45,  1.78s/it]
 51%|█████     | 7524/14725 [3:50:56<3:35:52,  1.80s/it]
 51%|█████     | 7525/14725 [3:50:58<3:33:48,  1.78s/it]
 51%|█████     | 7526/14725 [3:51:00<3:34:01,  1.78s/it]
 51%|████�
+0: {'loss': 0.8267, 'grad_norm': 0.69727897895741, 'learning_rate': 2e-05, 'epoch': 0.51}
+0: ��     | 7527/14725 [3:51:02<3:33:00,  1.78s/it]
 51%|█████     | 7528/14725 [3:51:04<3:33:30,  1.78s/it]
 51%|█████     | 7529/14725 [3:51:05<3:34:01,  1.78s/it]
 51%|█████     | 7530/14725 [3:51:07<3:35:07,  1.79s/it]
                                                        

 51%|█████     | 7530/14725 [3:51:07<3:35:07,  1.79s/it]
 51%|█████     | 7531/14725 [3:51:09<3:34:01,  1.78s/it]
 51%|█████     | 7532/14725 [3:51:11<3:40:17,  1.84s/it]
 51%|█████     | 7533/14725 [3:51:13<3:41:59,  1.85s/it]
 51%|█████     | 7534/14725 [3:51:15<3:40:11,  1.84s/it]
 51%|█████     | 7535/14725 [3:51:16<3:39:03,  1.83s/it]
 51%|█████     | 7536/14725 [3:51:18<3:36:55,  1.81s/it]
 51%|█████     | 7537/14725 [3:51:20<3:38:12,  1.82s/it]
 51%|█████     | 7538/14725 [3:51:22<3:35:42,  1.80s/it]
 51%|█████     | 7539/14725 [3:51:24<3:37:19,  1.81s/it]
 51%|█████     | 7540/14725 [3:51:25
+0: {'loss': 0.7929, 'grad_norm': 0.7039403088674673, 'learning_rate': 2e-05, 'epoch': 0.51}
+0: {'loss': 0.8151, 'grad_norm': 0.7290401047975075, 'learning_rate': 2e-05, 'epoch': 0.51}
+0: <3:36:25,  1.81s/it]
                                                        

 51%|█████     | 7540/14725 [3:51:25<3:36:25,  1.81s/it]
 51%|█████     | 7541/14725 [3:51:27<3:35:21,  1.80s/it]
 51%|█████     | 7542/14725 [3:51:29<3:34:52,  1.79s/it]
 51%|█████     | 7543/14725 [3:51:31<3:35:26,  1.80s/it]
 51%|█████     | 7544/14725 [3:51:33<3:36:13,  1.81s/it]
 51%|█████     | 7545/14725 [3:51:34<3:36:38,  1.81s/it]
 51%|█████     | 7546/14725 [3:51:36<3:38:17,  1.82s/it]
 51%|█████▏    | 7547/14725 [3:51:38<3:40:59,  1.85s/it]
 51%|█████▏    | 7548/14725 [3:51:40<3:42:02,  1.86s/it]
 51%|█████▏    | 7549/14725 [3:51:42<3:39:22,  1.83s/it]
 51%|█████▏    | 7550/14725 [3:51:44<3:40:48,  1.85s/it]
                                                        

 51%|█████▏    | 7550/14725 [3:51:44<3:40:48,  1.85s/it]
 51%|█████▏    | 7551/14725 [3:51:46<3:40:53,  1.85s/it]
 51%
+0: {'loss': 0.8313, 'grad_norm': 0.7148536523198289, 'learning_rate': 2e-05, 'epoch': 0.51}
+0: |█████▏    | 7552/14725 [3:51:47<3:39:25,  1.84s/it]
 51%|█████▏    | 7553/14725 [3:51:49<3:37:54,  1.82s/it]
 51%|█████▏    | 7554/14725 [3:51:51<3:35:30,  1.80s/it]
 51%|█████▏    | 7555/14725 [3:51:53<3:35:27,  1.80s/it]
 51%|█████▏    | 7556/14725 [3:51:55<3:34:41,  1.80s/it]
 51%|█████▏    | 7557/14725 [3:51:56<3:33:57,  1.79s/it]
 51%|█████▏    | 7558/14725 [3:51:58<3:34:21,  1.79s/it]
 51%|█████▏    | 7559/14725 [3:52:00<3:34:57,  1.80s/it]
 51%|█████▏    | 7560/14725 [3:52:02<3:33:26,  1.79s/it]
                                                        

 51%|█████▏    | 7560/14725 [3:52:02<3:33:26,  1.79s/it]
 51%|█████▏    | 7561/14725 [3:52:04<3:35:45,  1.81s/it]
 51%|█████▏    | 7562/14725 [3:52:05<3:35:11,  1.80s/it]
 51%|█████▏    | 7563/14725 [3:52:07<3:35:50,  1.81s/it]
 51%|█████▏    | 7564/14725 [3:52:09<3:39:26,  1.84s/it]
 51%
+0: {'loss': 0.8329, 'grad_norm': 0.7224538211017871, 'learning_rate': 2e-05, 'epoch': 0.51}
+0: |█████▏    | 7565/14725 [3:52:11<3:37:09,  1.82s/it]
 51%|█████▏    | 7566/14725 [3:52:13<3:38:14,  1.83s/it]
 51%|█████▏    | 7567/14725 [3:52:14<3:38:01,  1.83s/it]
 51%|█████▏    | 7568/14725 [3:52:16<3:36:37,  1.82s/it]
 51%|█████▏    | 7569/14725 [3:52:18<3:38:14,  1.83s/it]
 51%|█████▏    | 7570/14725 [3:52:20<3:36:58,  1.82s/it]
                                                        

 51%|█████▏    | 7570/14725 [3:52:20<3:36:58,  1.82s/it]
 51%|█████▏    | 7571/14725 [3:52:22<3:40:33,  1.85s/it]
 51%|█████▏    | 7572/14725 [3:52:24<3:37:39,  1.83s/it]
 51%|█████▏    | 7573/14725 [3:52:26<3:39:40,  1.84s/it]
 51%|█████▏    | 7574/14725 [3:52:27<3:37:22,  1.82s/it]
 51%|█████▏    | 7575/14725 [3:52:29<3:34:19,  1.80s/it]
 51%|█████▏    | 7576/14725 [3:52:31<3:34:58,  1.80s/it]
 51%|█████▏    | 7577/14725 [3:52:33<3:37:22,  1.82s/it]
 51%
+0: {'loss': 0.8114, 'grad_norm': 0.7828623572849829, 'learning_rate': 2e-05, 'epoch': 0.51}
+0: |█████▏    | 7578/14725 [3:52:35<3:37:21,  1.82s/it]
 51%|█████▏    | 7579/14725 [3:52:36<3:34:43,  1.80s/it]
 51%|█████▏    | 7580/14725 [3:52:38<3:34:46,  1.80s/it]
                                                        

 51%|█████▏    | 7580/14725 [3:52:38<3:34:46,  1.80s/it]
 51%|█████▏    | 7581/14725 [3:52:40<3:39:24,  1.84s/it]
 51%|█████▏    | 7582/14725 [3:52:42<3:39:05,  1.84s/it]
 51%|█████▏    | 7583/14725 [3:52:44<3:36:37,  1.82s/it]
 52%|█████▏    | 7584/14725 [3:52:45<3:36:34,  1.82s/it]
 52%|█████▏    | 7585/14725 [3:52:47<3:35:16,  1.81s/it]
 52%|█████▏    | 7586/14725 [3:52:49<3:34:47,  1.81s/it]
 52%|█████▏    | 7587/14725 [3:52:51<3:36:03,  1.82s/it]
 52%|█████▏    | 7588/14725 [3:52:53<3:37:08,  1.83s/it]
 52%|█████▏    | 7589/14725 [3:52:55<3:36:36,  1.82s/it]
 52%|█████▏    | 7590/14725 [3:52:56<3:39:02,  1.84s/it]
    
+0: {'loss': 0.8177, 'grad_norm': 0.7249467515092297, 'learning_rate': 2e-05, 'epoch': 0.52}
+0: {'loss': 0.8113, 'grad_norm': 0.7144810332705345, 'learning_rate': 2e-05, 'epoch': 0.52}
+0:                                                     

 52%|█████▏    | 7590/14725 [3:52:56<3:39:02,  1.84s/it]
 52%|█████▏    | 7591/14725 [3:52:58<3:36:24,  1.82s/it]
 52%|█████▏    | 7592/14725 [3:53:00<3:37:35,  1.83s/it]
 52%|█████▏    | 7593/14725 [3:53:02<3:38:07,  1.84s/it]
 52%|█████▏    | 7594/14725 [3:53:04<3:35:40,  1.81s/it]
 52%|█████▏    | 7595/14725 [3:53:05<3:33:02,  1.79s/it]
 52%|█████▏    | 7596/14725 [3:53:07<3:35:22,  1.81s/it]
 52%|█████▏    | 7597/14725 [3:53:09<3:35:00,  1.81s/it]
 52%|█████▏    | 7598/14725 [3:53:11<3:33:16,  1.80s/it]
 52%|█████▏    | 7599/14725 [3:53:13<3:34:54,  1.81s/it]
 52%|█████▏    | 7600/14725 [3:53:14<3:34:07,  1.80s/it]
                                                        

 52%|█████▏    | 7600/14725 [3:53:14<3:34:07,  1.80s/it]
 52%|█████▏    | 7601/14725 [3:53:16<3:35:38,  1.82s/it]
 52%|███�
+0: {'loss': 0.813, 'grad_norm': 0.7059008380539535, 'learning_rate': 2e-05, 'epoch': 0.52}
+0: ��█▏    | 7602/14725 [3:53:18<3:32:58,  1.79s/it]
 52%|█████▏    | 7603/14725 [3:53:20<3:33:05,  1.80s/it]
 52%|█████▏    | 7604/14725 [3:53:22<3:31:08,  1.78s/it]
 52%|█████▏    | 7605/14725 [3:53:23<3:30:47,  1.78s/it]
 52%|█████▏    | 7606/14725 [3:53:25<3:32:50,  1.79s/it]
 52%|█████▏    | 7607/14725 [3:53:27<3:36:41,  1.83s/it]
 52%|█████▏    | 7608/14725 [3:53:29<3:37:28,  1.83s/it]
 52%|█████▏    | 7609/14725 [3:53:31<3:36:35,  1.83s/it]
 52%|█████▏    | 7610/14725 [3:53:33<3:36:45,  1.83s/it]
                                                        

 52%|█████▏    | 7610/14725 [3:53:33<3:36:45,  1.83s/it]
 52%|█████▏    | 7611/14725 [3:53:34<3:36:38,  1.83s/it]
 52%|█████▏    | 7612/14725 [3:53:36<3:35:35,  1.82s/it]
 52%|█████▏    | 7613/14725 [3:53:38<3:35:48,  1.82s/it]
 52%|█████▏    | 7614/14725 [3:53:40<3:34:32,  1.81s/it]
 52%|███�
+0: {'loss': 0.8316, 'grad_norm': 0.7536491893820548, 'learning_rate': 2e-05, 'epoch': 0.52}
+0: ��█▏    | 7615/14725 [3:53:42<3:35:33,  1.82s/it]
 52%|█████▏    | 7616/14725 [3:53:43<3:35:39,  1.82s/it]
 52%|█████▏    | 7617/14725 [3:53:45<3:37:27,  1.84s/it]
 52%|█████▏    | 7618/14725 [3:53:47<3:36:02,  1.82s/it]
 52%|█████▏    | 7619/14725 [3:53:49<3:35:19,  1.82s/it]
 52%|█████▏    | 7620/14725 [3:53:51<3:33:41,  1.80s/it]
                                                        

 52%|█████▏    | 7620/14725 [3:53:51<3:33:41,  1.80s/it]
 52%|█████▏    | 7621/14725 [3:53:53<3:34:40,  1.81s/it]
 52%|█████▏    | 7622/14725 [3:53:54<3:34:27,  1.81s/it]
 52%|█████▏    | 7623/14725 [3:53:56<3:33:08,  1.80s/it]
 52%|█████▏    | 7624/14725 [3:53:58<3:31:54,  1.79s/it]
 52%|█████▏    | 7625/14725 [3:54:00<3:30:50,  1.78s/it]
 52%|█████▏    | 7626/14725 [3:54:01<3:30:41,  1.78s/it]
 52%|█████▏    | 7627/14725 [3:54:03<3:31:01,  1.78s/it]
 52%|███�
+0: {'loss': 0.8082, 'grad_norm': 0.7235963084411763, 'learning_rate': 2e-05, 'epoch': 0.52}
+0: ��█▏    | 7628/14725 [3:54:05<3:32:21,  1.80s/it]
 52%|█████▏    | 7629/14725 [3:54:07<3:32:55,  1.80s/it]
 52%|█████▏    | 7630/14725 [3:54:09<3:33:46,  1.81s/it]
                                                        

 52%|█████▏    | 7630/14725 [3:54:09<3:33:46,  1.81s/it]
 52%|█████▏    | 7631/14725 [3:54:10<3:32:56,  1.80s/it]
 52%|█████▏    | 7632/14725 [3:54:12<3:31:17,  1.79s/it]
 52%|█████▏    | 7633/14725 [3:54:14<3:29:52,  1.78s/it]
 52%|█████▏    | 7634/14725 [3:54:16<3:30:19,  1.78s/it]
 52%|█████▏    | 7635/14725 [3:54:18<3:33:22,  1.81s/it]
 52%|█████▏    | 7636/14725 [3:54:19<3:32:25,  1.80s/it]
 52%|█████▏    | 7637/14725 [3:54:21<3:30:45,  1.78s/it]
 52%|█████▏    | 7638/14725 [3:54:23<3:35:28,  1.82s/it]
 52%|█████▏    | 7639/14725 [3:54:25<3:34:48,  1.82s/it]
 52%|█████▏    | 7640/14725 [3:54:27<3:32:44,  1.80s/it]
               
+0: {'loss': 0.8299, 'grad_norm': 0.7360369105983066, 'learning_rate': 2e-05, 'epoch': 0.52}
+0: {'loss': 0.8068, 'grad_norm': 0.6781417638016012, 'learning_rate': 2e-05, 'epoch': 0.52}
+0:                                          

 52%|█████▏    | 7640/14725 [3:54:27<3:32:44,  1.80s/it]
 52%|█████▏    | 7641/14725 [3:54:28<3:33:15,  1.81s/it]
 52%|█████▏    | 7642/14725 [3:54:30<3:34:47,  1.82s/it]
 52%|█████▏    | 7643/14725 [3:54:32<3:35:09,  1.82s/it]
 52%|█████▏    | 7644/14725 [3:54:34<3:32:55,  1.80s/it]
 52%|█████▏    | 7645/14725 [3:54:36<3:40:41,  1.87s/it]
 52%|█████▏    | 7646/14725 [3:54:38<3:38:25,  1.85s/it]
 52%|█████▏    | 7647/14725 [3:54:40<3:37:47,  1.85s/it]
 52%|█████▏    | 7648/14725 [3:54:41<3:36:03,  1.83s/it]
 52%|█████▏    | 7649/14725 [3:54:43<3:33:52,  1.81s/it]
 52%|█████▏    | 7650/14725 [3:54:45<3:31:18,  1.79s/it]
                                                        

 52%|█████▏    | 7650/14725 [3:54:45<3:31:18,  1.79s/it]
 52%|█████▏    | 7651/14725 [3:54:47<3:31:09,  1.79s/it]
 52%|█████▏   
+0: {'loss': 0.8521, 'grad_norm': 0.745534563148455, 'learning_rate': 2e-05, 'epoch': 0.52}
+0:  | 7652/14725 [3:54:48<3:30:22,  1.78s/it]
 52%|█████▏    | 7653/14725 [3:54:50<3:29:39,  1.78s/it]
 52%|█████▏    | 7654/14725 [3:54:52<3:32:52,  1.81s/it]
 52%|█████▏    | 7655/14725 [3:54:54<3:35:54,  1.83s/it]
 52%|█████▏    | 7656/14725 [3:54:56<3:35:02,  1.83s/it]
 52%|█████▏    | 7657/14725 [3:54:58<3:33:02,  1.81s/it]
 52%|█████▏    | 7658/14725 [3:54:59<3:36:53,  1.84s/it]
 52%|█████▏    | 7659/14725 [3:55:01<3:38:24,  1.85s/it]
 52%|█████▏    | 7660/14725 [3:55:03<3:35:31,  1.83s/it]
                                                        

 52%|█████▏    | 7660/14725 [3:55:03<3:35:31,  1.83s/it]
 52%|█████▏    | 7661/14725 [3:55:05<3:36:59,  1.84s/it]
 52%|█████▏    | 7662/14725 [3:55:07<3:34:11,  1.82s/it]
 52%|█████▏    | 7663/14725 [3:55:09<3:31:54,  1.80s/it]
 52%|█████▏    | 7664/14725 [3:55:10<3:31:21,  1.80s/it]
 52%|█████▏   
+0: {'loss': 0.82, 'grad_norm': 0.7350436507548568, 'learning_rate': 2e-05, 'epoch': 0.52}
+0:  | 7665/14725 [3:55:12<3:30:25,  1.79s/it]
 52%|█████▏    | 7666/14725 [3:55:14<3:28:54,  1.78s/it]
 52%|█████▏    | 7667/14725 [3:55:16<3:29:42,  1.78s/it]
 52%|█████▏    | 7668/14725 [3:55:17<3:30:37,  1.79s/it]
 52%|█████▏    | 7669/14725 [3:55:19<3:30:02,  1.79s/it]
 52%|█████▏    | 7670/14725 [3:55:21<3:29:48,  1.78s/it]
                                                        

 52%|█████▏    | 7670/14725 [3:55:21<3:29:48,  1.78s/it]
 52%|█████▏    | 7671/14725 [3:55:23<3:30:24,  1.79s/it]
 52%|█████▏    | 7672/14725 [3:55:25<3:31:35,  1.80s/it]
 52%|█████▏    | 7673/14725 [3:55:26<3:31:58,  1.80s/it]
 52%|█████▏    | 7674/14725 [3:55:28<3:33:44,  1.82s/it]
 52%|█████▏    | 7675/14725 [3:55:30<3:30:58,  1.80s/it]
 52%|█████▏    | 7676/14725 [3:55:32<3:30:57,  1.80s/it]
 52%|█████▏    | 7677/14725 [3:55:34<3:31:25,  1.80s/it]
 52%|█████▏   
+0: {'loss': 0.8018, 'grad_norm': 0.6845184693840796, 'learning_rate': 2e-05, 'epoch': 0.52}
+0:  | 7678/14725 [3:55:35<3:30:47,  1.79s/it]
 52%|█████▏    | 7679/14725 [3:55:37<3:31:18,  1.80s/it]
 52%|█████▏    | 7680/14725 [3:55:39<3:31:58,  1.81s/it]
                                                        

 52%|█████▏    | 7680/14725 [3:55:39<3:31:58,  1.81s/it]
 52%|█████▏    | 7681/14725 [3:55:41<3:31:20,  1.80s/it]
 52%|█████▏    | 7682/14725 [3:55:43<3:31:33,  1.80s/it]
 52%|█████▏    | 7683/14725 [3:55:44<3:30:40,  1.79s/it]
 52%|█████▏    | 7684/14725 [3:55:46<3:30:47,  1.80s/it]
 52%|█████▏    | 7685/14725 [3:55:48<3:30:32,  1.79s/it]
 52%|█████▏    | 7686/14725 [3:55:50<3:28:58,  1.78s/it]
 52%|█████▏    | 7687/14725 [3:55:52<3:30:17,  1.79s/it]
 52%|█████▏    | 7688/14725 [3:55:53<3:31:02,  1.80s/it]
 52%|█████▏    | 7689/14725 [3:55:55<3:33:10,  1.82s/it]
 52%|█████▏    | 7690/14725 [3:55:57<3:31:41,  1.81s/it]
                          
+0: {'loss': 0.8239, 'grad_norm': 0.7057036560030243, 'learning_rate': 2e-05, 'epoch': 0.52}
+0: {'loss': 0.8231, 'grad_norm': 0.7088165478573715, 'learning_rate': 2e-05, 'epoch': 0.52}
+0:                               

 52%|█████▏    | 7690/14725 [3:55:57<3:31:41,  1.81s/it]
 52%|█████▏    | 7691/14725 [3:55:59<3:30:15,  1.79s/it]
 52%|█████▏    | 7692/14725 [3:56:01<3:33:21,  1.82s/it]
 52%|█████▏    | 7693/14725 [3:56:02<3:32:05,  1.81s/it]
 52%|█████▏    | 7694/14725 [3:56:04<3:29:52,  1.79s/it]
 52%|█████▏    | 7695/14725 [3:56:06<3:28:42,  1.78s/it]
 52%|█████▏    | 7696/14725 [3:56:08<3:28:24,  1.78s/it]
 52%|█████▏    | 7697/14725 [3:56:10<3:29:19,  1.79s/it]
 52%|█████▏    | 7698/14725 [3:56:11<3:29:00,  1.78s/it]
 52%|█████▏    | 7699/14725 [3:56:13<3:32:41,  1.82s/it]
 52%|█████▏    | 7700/14725 [3:56:15<3:31:58,  1.81s/it]
                                                        

 52%|█████▏    | 7700/14725 [3:56:15<3:31:58,  1.81s/it]
 52%|█████▏    | 7701/14725 [3:56:17<3:31:20,  1.81s/it]
 52%|█████▏    | 7702/147
+0: {'loss': 0.84, 'grad_norm': 0.7509540422303511, 'learning_rate': 2e-05, 'epoch': 0.52}
+0: 25 [3:56:19<3:31:00,  1.80s/it]
 52%|█████▏    | 7703/14725 [3:56:20<3:32:10,  1.81s/it]
 52%|█████▏    | 7704/14725 [3:56:22<3:32:25,  1.82s/it]
 52%|█████▏    | 7705/14725 [3:56:24<3:30:42,  1.80s/it]
 52%|█████▏    | 7706/14725 [3:56:26<3:31:11,  1.81s/it]
 52%|█████▏    | 7707/14725 [3:56:28<3:29:42,  1.79s/it]
 52%|█████▏    | 7708/14725 [3:56:29<3:28:55,  1.79s/it]
 52%|█████▏    | 7709/14725 [3:56:31<3:29:55,  1.80s/it]
 52%|█████▏    | 7710/14725 [3:56:33<3:28:03,  1.78s/it]
                                                        

 52%|█████▏    | 7710/14725 [3:56:33<3:28:03,  1.78s/it]
 52%|█████▏    | 7711/14725 [3:56:35<3:30:29,  1.80s/it]
 52%|█████▏    | 7712/14725 [3:56:37<3:32:26,  1.82s/it]
 52%|█████▏    | 7713/14725 [3:56:38<3:30:10,  1.80s/it]
 52%|█████▏    | 7714/14725 [3:56:40<3:31:48,  1.81s/it]
 52%|█████▏    | 7715/147
+0: {'loss': 0.8182, 'grad_norm': 0.6960551378263791, 'learning_rate': 2e-05, 'epoch': 0.52}
+0: 25 [3:56:42<3:36:01,  1.85s/it]
 52%|█████▏    | 7716/14725 [3:56:44<3:34:13,  1.83s/it]
 52%|█████▏    | 7717/14725 [3:56:46<3:34:35,  1.84s/it]
 52%|█████▏    | 7718/14725 [3:56:48<3:32:33,  1.82s/it]
 52%|█████▏    | 7719/14725 [3:56:49<3:30:04,  1.80s/it]
 52%|█████▏    | 7720/14725 [3:56:51<3:27:57,  1.78s/it]
                                                        

 52%|█████▏    | 7720/14725 [3:56:51<3:27:57,  1.78s/it]
 52%|█████▏    | 7721/14725 [3:56:53<3:30:02,  1.80s/it]
 52%|█████▏    | 7722/14725 [3:56:55<3:29:31,  1.80s/it]
 52%|█████▏    | 7723/14725 [3:56:57<3:33:26,  1.83s/it]
 52%|█████▏    | 7724/14725 [3:56:58<3:32:36,  1.82s/it]
 52%|█████▏    | 7725/14725 [3:57:00<3:31:54,  1.82s/it]
 52%|█████▏    | 7726/14725 [3:57:02<3:34:12,  1.84s/it]
 52%|█████▏    | 7727/14725 [3:57:04<3:31:06,  1.81s/it]
 52%|█████▏    | 7728/147
+0: {'loss': 0.812, 'grad_norm': 0.7485556777715235, 'learning_rate': 2e-05, 'epoch': 0.52}
+0: 25 [3:57:06<3:30:31,  1.81s/it]
 52%|█████▏    | 7729/14725 [3:57:08<3:36:37,  1.86s/it]
 52%|█████▏    | 7730/14725 [3:57:09<3:35:24,  1.85s/it]
                                                        

 52%|█████▏    | 7730/14725 [3:57:09<3:35:24,  1.85s/it]
 53%|█████▎    | 7731/14725 [3:57:11<3:33:45,  1.83s/it]
 53%|█████▎    | 7732/14725 [3:57:13<3:31:38,  1.82s/it]
 53%|█████▎    | 7733/14725 [3:57:15<3:31:36,  1.82s/it]
 53%|█████▎    | 7734/14725 [3:57:17<3:31:47,  1.82s/it]
 53%|█████▎    | 7735/14725 [3:57:18<3:30:15,  1.80s/it]
 53%|█████▎    | 7736/14725 [3:57:20<3:32:49,  1.83s/it]
 53%|█████▎    | 7737/14725 [3:57:22<3:30:33,  1.81s/it]
 53%|█████▎    | 7738/14725 [3:57:24<3:30:58,  1.81s/it]
 53%|█████▎    | 7739/14725 [3:57:26<3:30:06,  1.80s/it]
 53%|█████▎    | 7740/14725 [3:57:28<3:30:12,  1.81s/it]
                                     
+0: {'loss': 0.8227, 'grad_norm': 0.6728386140440974, 'learning_rate': 2e-05, 'epoch': 0.53}
+0: {'loss': 0.8216, 'grad_norm': 0.8088540793942673, 'learning_rate': 2e-05, 'epoch': 0.53}
+0:                    

 53%|█████▎    | 7740/14725 [3:57:28<3:30:12,  1.81s/it]
 53%|█████▎    | 7741/14725 [3:57:29<3:31:36,  1.82s/it]
 53%|█████▎    | 7742/14725 [3:57:31<3:30:21,  1.81s/it]
 53%|█████▎    | 7743/14725 [3:57:33<3:31:03,  1.81s/it]
 53%|█████▎    | 7744/14725 [3:57:35<3:30:48,  1.81s/it]
 53%|█████▎    | 7745/14725 [3:57:37<3:31:31,  1.82s/it]
 53%|█████▎    | 7746/14725 [3:57:38<3:31:02,  1.81s/it]
 53%|█████▎    | 7747/14725 [3:57:40<3:32:11,  1.82s/it]
 53%|█████▎    | 7748/14725 [3:57:42<3:30:27,  1.81s/it]
 53%|█████▎    | 7749/14725 [3:57:44<3:30:55,  1.81s/it]
 53%|█████▎    | 7750/14725 [3:57:46<3:29:31,  1.80s/it]
                                                        

 53%|█████▎    | 7750/14725 [3:57:46<3:29:31,  1.80s/it]
 53%|█████▎    | 7751/14725 [3:57:47<3:28:05,  1.79s/it]
 53%|█████▎    | 7752/14725 [3:57:49
+0: {'loss': 0.8282, 'grad_norm': 0.7863197687629316, 'learning_rate': 2e-05, 'epoch': 0.53}
+0: <3:29:05,  1.80s/it]
 53%|█████▎    | 7753/14725 [3:57:51<3:29:12,  1.80s/it]
 53%|█████▎    | 7754/14725 [3:57:53<3:31:51,  1.82s/it]
 53%|█████▎    | 7755/14725 [3:57:55<3:31:18,  1.82s/it]
 53%|█████▎    | 7756/14725 [3:57:57<3:29:45,  1.81s/it]
 53%|█████▎    | 7757/14725 [3:57:58<3:32:49,  1.83s/it]
 53%|█████▎    | 7758/14725 [3:58:00<3:29:49,  1.81s/it]
 53%|█████▎    | 7759/14725 [3:58:02<3:28:12,  1.79s/it]
 53%|█████▎    | 7760/14725 [3:58:04<3:29:09,  1.80s/it]
                                                        

 53%|█████▎    | 7760/14725 [3:58:04<3:29:09,  1.80s/it]
 53%|█████▎    | 7761/14725 [3:58:05<3:28:03,  1.79s/it]
 53%|█████▎    | 7762/14725 [3:58:07<3:28:44,  1.80s/it]
 53%|█████▎    | 7763/14725 [3:58:09<3:29:38,  1.81s/it]
 53%|█████▎    | 7764/14725 [3:58:11<3:28:00,  1.79s/it]
 53%|█████▎    | 7765/14725 [3:58:13
+0: {'loss': 0.8193, 'grad_norm': 0.7615350611304711, 'learning_rate': 2e-05, 'epoch': 0.53}
+0: <3:27:18,  1.79s/it]
 53%|█████▎    | 7766/14725 [3:58:14<3:28:18,  1.80s/it]
 53%|█████▎    | 7767/14725 [3:58:16<3:31:26,  1.82s/it]
 53%|█████▎    | 7768/14725 [3:58:18<3:29:35,  1.81s/it]
 53%|█████▎    | 7769/14725 [3:58:20<3:28:23,  1.80s/it]
 53%|█████▎    | 7770/14725 [3:58:22<3:28:48,  1.80s/it]
                                                        

 53%|█████▎    | 7770/14725 [3:58:22<3:28:48,  1.80s/it]
 53%|█████▎    | 7771/14725 [3:58:24<3:27:46,  1.79s/it]
 53%|█████▎    | 7772/14725 [3:58:25<3:27:31,  1.79s/it]
 53%|█████▎    | 7773/14725 [3:58:27<3:28:32,  1.80s/it]
 53%|█████▎    | 7774/14725 [3:58:29<3:30:11,  1.81s/it]
 53%|█████▎    | 7775/14725 [3:58:31<3:29:21,  1.81s/it]
 53%|█████▎    | 7776/14725 [3:58:33<3:29:10,  1.81s/it]
 53%|█████▎    | 7777/14725 [3:58:34<3:30:47,  1.82s/it]
 53%|█████▎    | 7778/14725 [3:58:36
+0: {'loss': 0.8372, 'grad_norm': 0.8043841293697498, 'learning_rate': 2e-05, 'epoch': 0.53}
+0: <3:28:37,  1.80s/it]
 53%|█████▎    | 7779/14725 [3:58:38<3:28:59,  1.81s/it]
 53%|█████▎    | 7780/14725 [3:58:40<3:44:28,  1.94s/it]
                                                        

 53%|█████▎    | 7780/14725 [3:58:40<3:44:28,  1.94s/it]
 53%|█████▎    | 7781/14725 [3:58:42<3:38:32,  1.89s/it]
 53%|█████▎    | 7782/14725 [3:58:44<3:36:35,  1.87s/it]
 53%|█████▎    | 7783/14725 [3:58:46<3:34:38,  1.86s/it]
 53%|█████▎    | 7784/14725 [3:58:48<3:40:07,  1.90s/it]
 53%|█████▎    | 7785/14725 [3:58:49<3:35:05,  1.86s/it]
 53%|█████▎    | 7786/14725 [3:58:51<3:32:30,  1.84s/it]
 53%|█████▎    | 7787/14725 [3:58:53<3:30:37,  1.82s/it]
 53%|█████▎    | 7788/14725 [3:58:55<3:28:17,  1.80s/it]
 53%|█████▎    | 7789/14725 [3:58:57<3:29:11,  1.81s/it]
 53%|█████▎    | 7790/14725 [3:58:58<3:29:03,  1.81s/it]
                                                
+0: {'loss': 0.8207, 'grad_norm': 0.697785455141222, 'learning_rate': 2e-05, 'epoch': 0.53}
+0: {'loss': 0.8012, 'grad_norm': 0.69025244018802, 'learning_rate': 2e-05, 'epoch': 0.53}
+0:         

 53%|█████▎    | 7790/14725 [3:58:58<3:29:03,  1.81s/it]
 53%|█████▎    | 7791/14725 [3:59:00<3:28:11,  1.80s/it]
 53%|█████▎    | 7792/14725 [3:59:02<3:27:31,  1.80s/it]
 53%|█████▎    | 7793/14725 [3:59:04<3:28:15,  1.80s/it]
 53%|█████▎    | 7794/14725 [3:59:06<3:31:47,  1.83s/it]
 53%|█████▎    | 7795/14725 [3:59:07<3:28:54,  1.81s/it]
 53%|█████▎    | 7796/14725 [3:59:09<3:30:05,  1.82s/it]
 53%|█████▎    | 7797/14725 [3:59:11<3:27:37,  1.80s/it]
 53%|█████▎    | 7798/14725 [3:59:13<3:28:41,  1.81s/it]
 53%|█████▎    | 7799/14725 [3:59:15<3:26:24,  1.79s/it]
 53%|█████▎    | 7800/14725 [3:59:17<3:36:30,  1.88s/it]
                                                        

 53%|█████▎    | 7800/14725 [3:59:17<3:36:30,  1.88s/it]
 53%|█████▎    | 7801/14725 [3:59:19<3:35:14,  1.87s/it]
 53%|█████▎    | 7802/14725 [3:59:20<3:32:05,  
+0: {'loss': 0.8414, 'grad_norm': 0.7212290754036472, 'learning_rate': 2e-05, 'epoch': 0.53}
+0: 1.84s/it]
 53%|█████▎    | 7803/14725 [3:59:22<3:30:58,  1.83s/it]
 53%|█████▎    | 7804/14725 [3:59:24<3:29:18,  1.81s/it]
 53%|█████▎    | 7805/14725 [3:59:26<3:27:17,  1.80s/it]
 53%|█████▎    | 7806/14725 [3:59:27<3:27:12,  1.80s/it]
 53%|█████▎    | 7807/14725 [3:59:29<3:29:32,  1.82s/it]
 53%|█████▎    | 7808/14725 [3:59:31<3:41:19,  1.92s/it]
 53%|█████▎    | 7809/14725 [3:59:33<3:39:47,  1.91s/it]
 53%|█████▎    | 7810/14725 [3:59:35<3:36:42,  1.88s/it]
                                                        

 53%|█████▎    | 7810/14725 [3:59:35<3:36:42,  1.88s/it]
 53%|█████▎    | 7811/14725 [3:59:37<3:34:46,  1.86s/it]
 53%|█████▎    | 7812/14725 [3:59:39<3:32:41,  1.85s/it]
 53%|█████▎    | 7813/14725 [3:59:41<3:29:32,  1.82s/it]
 53%|█████▎    | 7814/14725 [3:59:42<3:26:41,  1.79s/it]
 53%|█████▎    | 7815/14725 [3:59:44<3:28:29,  
+0: {'loss': 0.8273, 'grad_norm': 0.7362744702825337, 'learning_rate': 2e-05, 'epoch': 0.53}
+0: 1.81s/it]
 53%|█████▎    | 7816/14725 [3:59:46<3:29:09,  1.82s/it]
 53%|█████▎    | 7817/14725 [3:59:48<3:31:15,  1.83s/it]
 53%|█████▎    | 7818/14725 [3:59:50<3:29:01,  1.82s/it]
 53%|█████▎    | 7819/14725 [3:59:51<3:27:48,  1.81s/it]
 53%|█████▎    | 7820/14725 [3:59:53<3:28:01,  1.81s/it]
                                                        

 53%|█████▎    | 7820/14725 [3:59:53<3:28:01,  1.81s/it]
 53%|█████▎    | 7821/14725 [3:59:55<3:27:50,  1.81s/it]
 53%|█████▎    | 7822/14725 [3:59:57<3:27:11,  1.80s/it]
 53%|█████▎    | 7823/14725 [3:59:59<3:27:33,  1.80s/it]
 53%|█████▎    | 7824/14725 [4:00:00<3:28:37,  1.81s/it]
 53%|█████▎    | 7825/14725 [4:00:02<3:26:21,  1.79s/it]
 53%|█████▎    | 7826/14725 [4:00:04<3:28:07,  1.81s/it]
 53%|█████▎    | 7827/14725 [4:00:06<3:26:29,  1.80s/it]
 53%|█████▎    | 7828/14725 [4:00:08<3:25:17,  
+0: {'loss': 0.8277, 'grad_norm': 0.775704634151167, 'learning_rate': 2e-05, 'epoch': 0.53}
+0: {'loss': 0.812, 'grad_norm': 0.7463051074726014, 'learning_rate': 2e-05, 'epoch': 0.53}
+0: 1.79s/it]
 53%|█████▎    | 7829/14725 [4:00:09<3:27:40,  1.81s/it]
 53%|█████▎    | 7830/14725 [4:00:11<3:27:34,  1.81s/it]
                                                        

 53%|█████▎    | 7830/14725 [4:00:11<3:27:34,  1.81s/it]
 53%|█████▎    | 7831/14725 [4:00:13<3:38:54,  1.91s/it]
 53%|█████▎    | 7832/14725 [4:00:15<3:34:50,  1.87s/it]
 53%|█████��    | 7833/14725 [4:00:17<3:32:54,  1.85s/it]
 53%|█████▎    | 7834/14725 [4:00:19<3:31:23,  1.84s/it]
 53%|█████▎    | 7835/14725 [4:00:21<3:29:11,  1.82s/it]
 53%|█████▎    | 7836/14725 [4:00:22<3:27:57,  1.81s/it]
 53%|█████▎    | 7837/14725 [4:00:24<3:29:03,  1.82s/it]
 53%|█████▎    | 7838/14725 [4:00:26<3:28:42,  1.82s/it]
 53%|█████▎    | 7839/14725 [4:00:28<3:28:41,  1.82s/it]
 53%|█████▎    | 7840/14725 [4:00:30<3:28:21,  1.82s/it]
                                                        

 
+0: {'loss': 0.8048, 'grad_norm': 0.6797441677933082, 'learning_rate': 2e-05, 'epoch': 0.53}
+0: 53%|█████▎    | 7840/14725 [4:00:30<3:28:21,  1.82s/it]
 53%|█████▎    | 7841/14725 [4:00:31<3:27:18,  1.81s/it]
 53%|█████▎    | 7842/14725 [4:00:33<3:29:17,  1.82s/it]
 53%|█████▎    | 7843/14725 [4:00:35<3:29:25,  1.83s/it]
 53%|█████▎    | 7844/14725 [4:00:37<3:27:03,  1.81s/it]
 53%|█████▎    | 7845/14725 [4:00:39<3:31:45,  1.85s/it]
 53%|█████▎    | 7846/14725 [4:00:41<3:31:50,  1.85s/it]
 53%|█████▎    | 7847/14725 [4:00:42<3:32:01,  1.85s/it]
 53%|█████▎    | 7848/14725 [4:00:44<3:30:57,  1.84s/it]
 53%|█████▎    | 7849/14725 [4:00:46<3:29:36,  1.83s/it]
 53%|█████▎    | 7850/14725 [4:00:48<3:27:04,  1.81s/it]
                                                        

 53%|█████▎    | 7850/14725 [4:00:48<3:27:04,  1.81s/it]
 53%|█████▎    | 7851/14725 [4:00:50<3:26:06,  1.80s/it]
 53%|█████▎    | 7852/14725 [4:00:51<3:25:27,  1.79s/it]
 
+0: {'loss': 0.8207, 'grad_norm': 0.7235080948947435, 'learning_rate': 2e-05, 'epoch': 0.53}
+0: 53%|█████▎    | 7853/14725 [4:00:53<3:24:48,  1.79s/it]
 53%|█████▎    | 7854/14725 [4:00:55<3:24:00,  1.78s/it]
 53%|█████▎    | 7855/14725 [4:00:57<3:24:03,  1.78s/it]
 53%|█████▎    | 7856/14725 [4:00:59<3:26:58,  1.81s/it]
 53%|█████▎    | 7857/14725 [4:01:00<3:27:10,  1.81s/it]
 53%|█████▎    | 7858/14725 [4:01:02<3:26:36,  1.81s/it]
 53%|█████▎    | 7859/14725 [4:01:04<3:28:38,  1.82s/it]
 53%|█████▎    | 7860/14725 [4:01:06<3:26:43,  1.81s/it]
                                                        

 53%|█████▎    | 7860/14725 [4:01:06<3:26:43,  1.81s/it]
 53%|█████▎    | 7861/14725 [4:01:08<3:26:43,  1.81s/it]
 53%|█████▎    | 7862/14725 [4:01:09<3:26:18,  1.80s/it]
 53%|█████▎    | 7863/14725 [4:01:11<3:26:35,  1.81s/it]
 53%|█████▎    | 7864/14725 [4:01:13<3:28:04,  1.82s/it]
 53%|█████▎    | 7865/14725 [4:01:15<3:27:02,  1.81s/it]
 
+0: {'loss': 0.8016, 'grad_norm': 0.7060054984447212, 'learning_rate': 2e-05, 'epoch': 0.53}
+0: 53%|█████▎    | 7866/14725 [4:01:17<3:25:53,  1.80s/it]
 53%|█████▎    | 7867/14725 [4:01:19<3:27:44,  1.82s/it]
 53%|█████▎    | 7868/14725 [4:01:20<3:28:03,  1.82s/it]
 53%|█████▎    | 7869/14725 [4:01:22<3:31:32,  1.85s/it]
 53%|█████▎    | 7870/14725 [4:01:24<3:28:42,  1.83s/it]
                                                        

 53%|█████▎    | 7870/14725 [4:01:24<3:28:42,  1.83s/it]
 53%|█████▎    | 7871/14725 [4:01:26<3:28:07,  1.82s/it]
 53%|█████▎    | 7872/14725 [4:01:28<3:27:19,  1.82s/it]
 53%|█████▎    | 7873/14725 [4:01:29<3:26:58,  1.81s/it]
 53%|█████▎    | 7874/14725 [4:01:31<3:27:39,  1.82s/it]
 53%|█████▎    | 7875/14725 [4:01:33<3:27:07,  1.81s/it]
 53%|█████▎    | 7876/14725 [4:01:35<3:28:50,  1.83s/it]
 53%|█████▎    | 7877/14725 [4:01:37<3:30:54,  1.85s/it]
 54%|█████▎    | 7878/14725 [4:01:39<3:27:14,  1.82s/it]
 
+0: {'loss': 0.8233, 'grad_norm': 0.7040147828617659, 'learning_rate': 2e-05, 'epoch': 0.54}
+0: {'loss': 0.8377, 'grad_norm': 0.7288674359815672, 'learning_rate': 2e-05, 'epoch': 0.54}
+0: 54%|█████▎    | 7879/14725 [4:01:40<3:28:19,  1.83s/it]
 54%|█████▎    | 7880/14725 [4:01:42<3:25:46,  1.80s/it]
                                                        

 54%|█████▎    | 7880/14725 [4:01:42<3:25:46,  1.80s/it]
 54%|█████▎    | 7881/14725 [4:01:44<3:26:07,  1.81s/it]
 54%|█████▎    | 7882/14725 [4:01:46<3:28:18,  1.83s/it]
 54%|█████▎    | 7883/14725 [4:01:48<3:28:56,  1.83s/it]
 54%|█████▎    | 7884/14725 [4:01:50<3:30:41,  1.85s/it]
 54%|█████▎    | 7885/14725 [4:01:51<3:27:53,  1.82s/it]
 54%|█████▎    | 7886/14725 [4:01:53<3:25:56,  1.81s/it]
 54%|█████▎    | 7887/14725 [4:01:55<3:24:09,  1.79s/it]
 54%|█████▎    | 7888/14725 [4:01:57<3:24:42,  1.80s/it]
 54%|█████▎    | 7889/14725 [4:01:59<3:24:24,  1.79s/it]
 54%|█████▎    | 7890/14725 [4:02:00<3:25:07,  1.80s/it]
                                                        

 54%|██�
+0: {'loss': 0.7923, 'grad_norm': 0.7073553492816073, 'learning_rate': 2e-05, 'epoch': 0.54}
+0: ��██▎    | 7890/14725 [4:02:00<3:25:07,  1.80s/it]
 54%|█████▎    | 7891/14725 [4:02:02<3:25:52,  1.81s/it]
 54%|█████▎    | 7892/14725 [4:02:04<3:24:23,  1.79s/it]
 54%|█████▎    | 7893/14725 [4:02:06<3:24:17,  1.79s/it]
 54%|█████▎    | 7894/14725 [4:02:08<3:25:01,  1.80s/it]
 54%|█████▎    | 7895/14725 [4:02:09<3:26:20,  1.81s/it]
 54%|█████▎    | 7896/14725 [4:02:11<3:25:15,  1.80s/it]
 54%|█████▎    | 7897/14725 [4:02:13<3:23:32,  1.79s/it]
 54%|█████▎    | 7898/14725 [4:02:15<3:25:11,  1.80s/it]
 54%|█████▎    | 7899/14725 [4:02:17<3:23:47,  1.79s/it]
 54%|█████▎    | 7900/14725 [4:02:18<3:24:58,  1.80s/it]
                                                        

 54%|█████▎    | 7900/14725 [4:02:18<3:24:58,  1.80s/it]
 54%|█████▎    | 7901/14725 [4:02:20<3:27:17,  1.82s/it]
 54%|█████▎    | 7902/14725 [4:02:22<3:26:40,  1.82s/it]
 54%|██�
+0: {'loss': 0.814, 'grad_norm': 0.7203969183174775, 'learning_rate': 2e-05, 'epoch': 0.54}
+0: ��██▎    | 7903/14725 [4:02:24<3:27:17,  1.82s/it]
 54%|█████▎    | 7904/14725 [4:02:26<3:28:21,  1.83s/it]
 54%|█████▎    | 7905/14725 [4:02:27<3:26:00,  1.81s/it]
 54%|█████▎    | 7906/14725 [4:02:29<3:28:17,  1.83s/it]
 54%|█████▎    | 7907/14725 [4:02:31<3:26:25,  1.82s/it]
 54%|█████▎    | 7908/14725 [4:02:33<3:26:53,  1.82s/it]
 54%|█████▎    | 7909/14725 [4:02:35<3:25:27,  1.81s/it]
 54%|█████▎    | 7910/14725 [4:02:37<3:23:55,  1.80s/it]
                                                        

 54%|█████▎    | 7910/14725 [4:02:37<3:23:55,  1.80s/it]
 54%|█████▎    | 7911/14725 [4:02:38<3:25:29,  1.81s/it]
 54%|█████▎    | 7912/14725 [4:02:40<3:23:25,  1.79s/it]
 54%|█████▎    | 7913/14725 [4:02:42<3:21:25,  1.77s/it]
 54%|█████▎    | 7914/14725 [4:02:44<3:23:30,  1.79s/it]
 54%|█████▍    | 7915/14725 [4:02:45<3:22:46,  1.79s/it]
 54%|██�
+0: {'loss': 0.829, 'grad_norm': 0.7226005496884182, 'learning_rate': 2e-05, 'epoch': 0.54}
+0: ��██▍    | 7916/14725 [4:02:47<3:21:45,  1.78s/it]
 54%|█████▍    | 7917/14725 [4:02:49<3:22:09,  1.78s/it]
 54%|█████▍    | 7918/14725 [4:02:51<3:24:22,  1.80s/it]
 54%|█████▍    | 7919/14725 [4:02:53<3:25:28,  1.81s/it]
 54%|█████▍    | 7920/14725 [4:02:55<3:27:18,  1.83s/it]
                                                        

 54%|█████▍    | 7920/14725 [4:02:55<3:27:18,  1.83s/it]
 54%|█████▍    | 7921/14725 [4:02:56<3:31:07,  1.86s/it]
 54%|█████▍    | 7922/14725 [4:02:58<3:29:04,  1.84s/it]
 54%|█████▍    | 7923/14725 [4:03:00<3:26:45,  1.82s/it]
 54%|█████▍    | 7924/14725 [4:03:02<3:25:45,  1.82s/it]
 54%|█████▍    | 7925/14725 [4:03:04<3:25:14,  1.81s/it]
 54%|█████▍    | 7926/14725 [4:03:05<3:23:28,  1.80s/it]
 54%|█████▍    | 7927/14725 [4:03:07<3:21:43,  1.78s/it]
 54%|█████▍    | 7928/14725 [4:03:09<3:24:35,  1.81s/it]
 54%|██�
+0: {'loss': 0.8299, 'grad_norm': 0.6955493326735813, 'learning_rate': 2e-05, 'epoch': 0.54}
+0: {'loss': 0.8072, 'grad_norm': 0.7398240556375082, 'learning_rate': 2e-05, 'epoch': 0.54}
+0: ��██▍    | 7929/14725 [4:03:11<3:22:54,  1.79s/it]
 54%|█████▍    | 7930/14725 [4:03:13<3:24:45,  1.81s/it]
                                                        

 54%|█████▍    | 7930/14725 [4:03:13<3:24:45,  1.81s/it]
 54%|█████▍    | 7931/14725 [4:03:14<3:23:39,  1.80s/it]
 54%|█████▍    | 7932/14725 [4:03:16<3:22:32,  1.79s/it]
 54%|█████▍    | 7933/14725 [4:03:18<3:23:38,  1.80s/it]
 54%|█████▍    | 7934/14725 [4:03:20<3:23:09,  1.79s/it]
 54%|█████▍    | 7935/14725 [4:03:22<3:24:12,  1.80s/it]
 54%|█████▍    | 7936/14725 [4:03:23<3:24:41,  1.81s/it]
 54%|█████▍    | 7937/14725 [4:03:25<3:23:01,  1.79s/it]
 54%|█████▍    | 7938/14725 [4:03:27<3:24:38,  1.81s/it]
 54%|█████▍    | 7939/14725 [4:03:29<3:24:16,  1.81s/it]
 54%|█████▍    | 7940/14725 [4:03:31<3:26:58,  1.83s/it]
                                                        

 54%|█████▍
+0: {'loss': 0.8126, 'grad_norm': 0.6760136862218166, 'learning_rate': 2e-05, 'epoch': 0.54}
+0:     | 7940/14725 [4:03:31<3:26:58,  1.83s/it]
 54%|█████▍    | 7941/14725 [4:03:33<3:27:19,  1.83s/it]
 54%|█████▍    | 7942/14725 [4:03:34<3:24:49,  1.81s/it]
 54%|█████▍    | 7943/14725 [4:03:36<3:22:57,  1.80s/it]
 54%|█████▍    | 7944/14725 [4:03:38<3:21:29,  1.78s/it]
 54%|█████▍    | 7945/14725 [4:03:40<3:23:53,  1.80s/it]
 54%|█████▍    | 7946/14725 [4:03:41<3:23:44,  1.80s/it]
 54%|█████▍    | 7947/14725 [4:03:43<3:22:40,  1.79s/it]
 54%|█████▍    | 7948/14725 [4:03:45<3:23:14,  1.80s/it]
 54%|█████▍    | 7949/14725 [4:03:47<3:21:58,  1.79s/it]
 54%|█████▍    | 7950/14725 [4:03:49<3:22:29,  1.79s/it]
                                                        

 54%|█████▍    | 7950/14725 [4:03:49<3:22:29,  1.79s/it]
 54%|█████▍    | 7951/14725 [4:03:50<3:22:45,  1.80s/it]
 54%|█████▍    | 7952/14725 [4:03:52<3:21:26,  1.78s/it]
 54%|█████▍
+0: {'loss': 0.8091, 'grad_norm': 0.7042119286763143, 'learning_rate': 2e-05, 'epoch': 0.54}
+0:     | 7953/14725 [4:03:54<3:23:10,  1.80s/it]
 54%|█████▍    | 7954/14725 [4:03:56<3:23:52,  1.81s/it]
 54%|█████▍    | 7955/14725 [4:03:58<3:24:31,  1.81s/it]
 54%|█████▍    | 7956/14725 [4:03:59<3:23:10,  1.80s/it]
 54%|█████▍    | 7957/14725 [4:04:01<3:22:08,  1.79s/it]
 54%|█████▍    | 7958/14725 [4:04:03<3:22:06,  1.79s/it]
 54%|█████▍    | 7959/14725 [4:04:05<3:21:43,  1.79s/it]
 54%|█████▍    | 7960/14725 [4:04:07<3:24:39,  1.82s/it]
                                                        

 54%|█████▍    | 7960/14725 [4:04:07<3:24:39,  1.82s/it]
 54%|█████▍    | 7961/14725 [4:04:09<3:25:28,  1.82s/it]
 54%|█████▍    | 7962/14725 [4:04:10<3:24:08,  1.81s/it]
 54%|█████▍    | 7963/14725 [4:04:12<3:24:49,  1.82s/it]
 54%|█████▍    | 7964/14725 [4:04:14<3:22:05,  1.79s/it]
 54%|█████▍    | 7965/14725 [4:04:16<3:21:10,  1.79s/it]
 54%|█████▍
+0: {'loss': 0.8118, 'grad_norm': 0.6745189201196666, 'learning_rate': 2e-05, 'epoch': 0.54}
+0:     | 7966/14725 [4:04:18<3:25:32,  1.82s/it]
 54%|█████▍    | 7967/14725 [4:04:19<3:24:09,  1.81s/it]
 54%|█████▍    | 7968/14725 [4:04:21<3:26:01,  1.83s/it]
 54%|█████▍    | 7969/14725 [4:04:23<3:26:33,  1.83s/it]
 54%|█████▍    | 7970/14725 [4:04:25<3:23:25,  1.81s/it]
                                                        

 54%|█████▍    | 7970/14725 [4:04:25<3:23:25,  1.81s/it]
 54%|█████▍    | 7971/14725 [4:04:27<3:26:01,  1.83s/it]
 54%|█████▍    | 7972/14725 [4:04:29<3:28:03,  1.85s/it]
 54%|█████▍    | 7973/14725 [4:04:30<3:24:31,  1.82s/it]
 54%|█████▍    | 7974/14725 [4:04:32<3:22:08,  1.80s/it]
 54%|█████▍    | 7975/14725 [4:04:34<3:21:15,  1.79s/it]
 54%|█████▍    | 7976/14725 [4:04:36<3:22:38,  1.80s/it]
 54%|█████▍    | 7977/14725 [4:04:37<3:22:16,  1.80s/it]
 54%|█████▍    | 7978/14725 [4:04:39<3:21:02,  1.79s/it]
 54%|█████▍
+0: {'loss': 0.8474, 'grad_norm': 0.7712553460828444, 'learning_rate': 2e-05, 'epoch': 0.54}
+0: {'loss': 0.8184, 'grad_norm': 0.6957074724483808, 'learning_rate': 2e-05, 'epoch': 0.54}
+0:     | 7979/14725 [4:04:41<3:21:43,  1.79s/it]
 54%|█████▍    | 7980/14725 [4:04:43<3:22:53,  1.80s/it]
                                                        

 54%|█████▍    | 7980/14725 [4:04:43<3:22:53,  1.80s/it]
 54%|█████▍    | 7981/14725 [4:04:45<3:21:31,  1.79s/it]
 54%|█████▍    | 7982/14725 [4:04:46<3:20:12,  1.78s/it]
 54%|█████▍    | 7983/14725 [4:04:48<3:21:27,  1.79s/it]
 54%|█████▍    | 7984/14725 [4:04:50<3:23:55,  1.82s/it]
 54%|█████▍    | 7985/14725 [4:04:52<3:21:55,  1.80s/it]
 54%|█████▍    | 7986/14725 [4:04:54<3:21:36,  1.79s/it]
 54%|█████▍    | 7987/14725 [4:04:55<3:22:51,  1.81s/it]
 54%|█████▍    | 7988/14725 [4:04:57<3:21:57,  1.80s/it]
 54%|█████▍    | 7989/14725 [4:04:59<3:22:39,  1.81s/it]
 54%|█████▍    | 7990/14725 [4:05:01<3:23:34,  1.81s/it]
                                                        

 54%|█████▍    | 7990/
+0: {'loss': 0.8162, 'grad_norm': 0.7189017286770737, 'learning_rate': 2e-05, 'epoch': 0.54}
+0: 14725 [4:05:01<3:23:34,  1.81s/it]
 54%|█████▍    | 7991/14725 [4:05:03<3:23:41,  1.81s/it]
 54%|█████▍    | 7992/14725 [4:05:04<3:22:20,  1.80s/it]
 54%|█████▍    | 7993/14725 [4:05:06<3:21:17,  1.79s/it]
 54%|█████▍    | 7994/14725 [4:05:08<3:22:31,  1.81s/it]
 54%|█████▍    | 7995/14725 [4:05:10<3:23:10,  1.81s/it]
 54%|█████▍    | 7996/14725 [4:05:12<3:21:59,  1.80s/it]
 54%|█████▍    | 7997/14725 [4:05:13<3:21:25,  1.80s/it]
 54%|█████▍    | 7998/14725 [4:05:15<3:20:45,  1.79s/it]
 54%|█████▍    | 7999/14725 [4:05:17<3:20:33,  1.79s/it]
 54%|█████▍    | 8000/14725 [4:05:19<3:22:09,  1.80s/it]
                                                        

 54%|█████▍    | 8000/14725 [4:05:19<3:22:09,  1.80s/it]
 54%|█████▍    | 8001/14725 [4:05:21<3:22:33,  1.81s/it]
 54%|█████▍    | 8002/14725 [4:05:23<3:22:35,  1.81s/it]
 54%|█████▍    | 8003/
+0: {'loss': 0.8151, 'grad_norm': 0.7186307429611953, 'learning_rate': 2e-05, 'epoch': 0.54}
+0: 14725 [4:05:24<3:24:26,  1.82s/it]
 54%|█████▍    | 8004/14725 [4:05:26<3:22:39,  1.81s/it]
 54%|█████▍    | 8005/14725 [4:05:28<3:22:26,  1.81s/it]
 54%|█████▍    | 8006/14725 [4:05:30<3:22:41,  1.81s/it]
 54%|█████▍    | 8007/14725 [4:05:32<3:22:07,  1.81s/it]
 54%|█████▍    | 8008/14725 [4:05:33<3:20:13,  1.79s/it]
 54%|█████▍    | 8009/14725 [4:05:35<3:19:05,  1.78s/it]
 54%|█████▍    | 8010/14725 [4:05:37<3:18:26,  1.77s/it]
                                                        

 54%|█████▍    | 8010/14725 [4:05:37<3:18:26,  1.77s/it]
 54%|█████▍    | 8011/14725 [4:05:39<3:22:59,  1.81s/it]
 54%|█████▍    | 8012/14725 [4:05:41<3:23:13,  1.82s/it]
 54%|█████▍    | 8013/14725 [4:05:42<3:23:28,  1.82s/it]
 54%|█████▍    | 8014/14725 [4:05:44<3:20:56,  1.80s/it]
 54%|█████▍    | 8015/14725 [4:05:46<3:19:36,  1.78s/it]
 54%|█████▍    | 8016/
+0: {'loss': 0.8269, 'grad_norm': 0.7175654896723914, 'learning_rate': 2e-05, 'epoch': 0.54}
+0: 14725 [4:05:48<3:20:49,  1.80s/it]
 54%|█████▍    | 8017/14725 [4:05:49<3:20:57,  1.80s/it]
 54%|█████▍    | 8018/14725 [4:05:51<3:20:10,  1.79s/it]
 54%|█████▍    | 8019/14725 [4:05:53<3:18:37,  1.78s/it]
 54%|█████▍    | 8020/14725 [4:05:55<3:22:10,  1.81s/it]
                                                        

 54%|█████▍    | 8020/14725 [4:05:55<3:22:10,  1.81s/it]
 54%|█████▍    | 8021/14725 [4:05:57<3:20:27,  1.79s/it]
 54%|█████▍    | 8022/14725 [4:05:58<3:18:26,  1.78s/it]
 54%|█████▍    | 8023/14725 [4:06:00<3:18:50,  1.78s/it]
 54%|█████▍    | 8024/14725 [4:06:02<3:20:15,  1.79s/it]
 54%|█████▍    | 8025/14725 [4:06:04<3:19:38,  1.79s/it]
 55%|█████▍    | 8026/14725 [4:06:06<3:19:54,  1.79s/it]
 55%|█████▍    | 8027/14725 [4:06:07<3:21:55,  1.81s/it]
 55%|█████▍    | 8028/14725 [4:06:09<3:20:30,  1.80s/it]
 55%|█████▍    | 8029/
+0: {'loss': 0.8067, 'grad_norm': 0.7097929607513734, 'learning_rate': 2e-05, 'epoch': 0.55}
+0: {'loss': 0.827, 'grad_norm': 0.7216158211302008, 'learning_rate': 2e-05, 'epoch': 0.55}
+0: 14725 [4:06:11<3:20:50,  1.80s/it]
 55%|█████▍    | 8030/14725 [4:06:13<3:22:01,  1.81s/it]
                                                        

 55%|█████▍    | 8030/14725 [4:06:13<3:22:01,  1.81s/it]
 55%|█████▍    | 8031/14725 [4:06:15<3:20:25,  1.80s/it]
 55%|█████▍    | 8032/14725 [4:06:16<3:20:41,  1.80s/it]
 55%|█████▍    | 8033/14725 [4:06:18<3:19:55,  1.79s/it]
 55%|█████▍    | 8034/14725 [4:06:20<3:23:34,  1.83s/it]
 55%|█████▍    | 8035/14725 [4:06:22<3:21:17,  1.81s/it]
 55%|█████▍    | 8036/14725 [4:06:24<3:25:34,  1.84s/it]
 55%|█████▍    | 8037/14725 [4:06:26<3:24:53,  1.84s/it]
 55%|█████▍    | 8038/14725 [4:06:27<3:21:40,  1.81s/it]
 55%|█████▍    | 8039/14725 [4:06:29<3:21:35,  1.81s/it]
 55%|█████▍    | 8040/14725 [4:06:31<3:33:38,  1.92s/it]
                                                        

 55%|█████▍    | 8040/14725 [4:06
+0: {'loss': 0.8008, 'grad_norm': 0.6913906420148518, 'learning_rate': 2e-05, 'epoch': 0.55}
+0: :31<3:33:38,  1.92s/it]
 55%|█████▍    | 8041/14725 [4:06:33<3:35:12,  1.93s/it]
 55%|█████▍    | 8042/14725 [4:06:35<3:31:42,  1.90s/it]
 55%|█████▍    | 8043/14725 [4:06:37<3:26:52,  1.86s/it]
 55%|█████▍    | 8044/14725 [4:06:39<3:27:15,  1.86s/it]
 55%|█████▍    | 8045/14725 [4:06:41<3:26:30,  1.85s/it]
 55%|█████▍    | 8046/14725 [4:06:42<3:25:32,  1.85s/it]
 55%|█████▍    | 8047/14725 [4:06:44<3:23:21,  1.83s/it]
 55%|█████▍    | 8048/14725 [4:06:46<3:21:32,  1.81s/it]
 55%|█████▍    | 8049/14725 [4:06:48<3:23:39,  1.83s/it]
 55%|█████▍    | 8050/14725 [4:06:50<3:21:32,  1.81s/it]
                                                        

 55%|█████▍    | 8050/14725 [4:06:50<3:21:32,  1.81s/it]
 55%|█████▍    | 8051/14725 [4:06:51<3:20:50,  1.81s/it]
 55%|█████▍    | 8052/14725 [4:06:53<3:21:20,  1.81s/it]
 55%|█████▍    | 8053/14725 [4:06
+0: {'loss': 0.8285, 'grad_norm': 0.6960920512831562, 'learning_rate': 2e-05, 'epoch': 0.55}
+0: :55<3:21:58,  1.82s/it]
 55%|█████▍    | 8054/14725 [4:06:57<3:24:17,  1.84s/it]
 55%|█████▍    | 8055/14725 [4:06:59<3:27:58,  1.87s/it]
 55%|█████▍    | 8056/14725 [4:07:01<3:24:39,  1.84s/it]
 55%|█████▍    | 8057/14725 [4:07:03<3:25:29,  1.85s/it]
 55%|█████▍    | 8058/14725 [4:07:04<3:23:04,  1.83s/it]
 55%|█████▍    | 8059/14725 [4:07:06<3:20:20,  1.80s/it]
 55%|█████▍    | 8060/14725 [4:07:08<3:20:22,  1.80s/it]
                                                        

 55%|█████▍    | 8060/14725 [4:07:08<3:20:22,  1.80s/it]
 55%|█████▍    | 8061/14725 [4:07:10<3:19:33,  1.80s/it]
 55%|█████▍    | 8062/14725 [4:07:11<3:17:58,  1.78s/it]
 55%|█████▍    | 8063/14725 [4:07:13<3:19:29,  1.80s/it]
 55%|█████▍    | 8064/14725 [4:07:15<3:19:03,  1.79s/it]
 55%|█████▍    | 8065/14725 [4:07:17<3:18:18,  1.79s/it]
 55%|█████▍    | 8066/14725 [4:07
+0: {'loss': 0.8165, 'grad_norm': 0.7051614858314504, 'learning_rate': 2e-05, 'epoch': 0.55}
+0: :19<3:18:58,  1.79s/it]
 55%|█████▍    | 8067/14725 [4:07:20<3:18:07,  1.79s/it]
 55%|█████▍    | 8068/14725 [4:07:22<3:20:27,  1.81s/it]
 55%|█████▍    | 8069/14725 [4:07:24<3:22:08,  1.82s/it]
 55%|█████▍    | 8070/14725 [4:07:26<3:21:08,  1.81s/it]
                                                        

 55%|█████▍    | 8070/14725 [4:07:26<3:21:08,  1.81s/it]
 55%|█████▍    | 8071/14725 [4:07:28<3:22:34,  1.83s/it]
 55%|█████▍    | 8072/14725 [4:07:29<3:20:30,  1.81s/it]
 55%|█████▍    | 8073/14725 [4:07:31<3:21:31,  1.82s/it]
 55%|█████▍    | 8074/14725 [4:07:33<3:21:10,  1.81s/it]
 55%|█████▍    | 8075/14725 [4:07:35<3:22:09,  1.82s/it]
 55%|█████▍    | 8076/14725 [4:07:37<3:21:37,  1.82s/it]
 55%|█████▍    | 8077/14725 [4:07:39<3:19:59,  1.80s/it]
 55%|█████▍    | 8078/14725 [4:07:40<3:18:55,  1.80s/it]
 55%|█████▍    | 8079/14725 [4:07
+0: {'loss': 0.8216, 'grad_norm': 0.7079208436226199, 'learning_rate': 2e-05, 'epoch': 0.55}
+0: {'loss': 0.8294, 'grad_norm': 0.7383957195544609, 'learning_rate': 2e-05, 'epoch': 0.55}
+0: :42<3:21:22,  1.82s/it]
 55%|█████▍    | 8080/14725 [4:07:44<3:19:14,  1.80s/it]
                                                        

 55%|█████▍    | 8080/14725 [4:07:44<3:19:14,  1.80s/it]
 55%|█████▍    | 8081/14725 [4:07:46<3:19:16,  1.80s/it]
 55%|█████▍    | 8082/14725 [4:07:48<3:18:23,  1.79s/it]
 55%|█████▍    | 8083/14725 [4:07:49<3:16:59,  1.78s/it]
 55%|█████▍    | 8084/14725 [4:07:51<3:16:00,  1.77s/it]
 55%|█████▍    | 8085/14725 [4:07:53<3:18:15,  1.79s/it]
 55%|█████▍    | 8086/14725 [4:07:55<3:21:08,  1.82s/it]
 55%|█████▍    | 8087/14725 [4:07:57<3:20:52,  1.82s/it]
 55%|█████▍    | 8088/14725 [4:07:58<3:19:07,  1.80s/it]
 55%|█████▍    | 8089/14725 [4:08:00<3:17:03,  1.78s/it]
 55%|█████▍    | 8090/14725 [4:08:02<3:17:51,  1.79s/it]
                                                        

 55%|█████▍    | 8090/14725 [4:08:02<3:17:51
+0: {'loss': 0.8334, 'grad_norm': 0.724081398528832, 'learning_rate': 2e-05, 'epoch': 0.55}
+0: ,  1.79s/it]
 55%|█████▍    | 8091/14725 [4:08:04<3:16:38,  1.78s/it]
 55%|█████▍    | 8092/14725 [4:08:06<3:20:11,  1.81s/it]
 55%|█████▍    | 8093/14725 [4:08:07<3:20:27,  1.81s/it]
 55%|█████▍    | 8094/14725 [4:08:09<3:19:15,  1.80s/it]
 55%|█████▍    | 8095/14725 [4:08:11<3:20:15,  1.81s/it]
 55%|█████▍    | 8096/14725 [4:08:13<3:20:42,  1.82s/it]
 55%|█████▍    | 8097/14725 [4:08:15<3:21:53,  1.83s/it]
 55%|█████▍    | 8098/14725 [4:08:16<3:21:07,  1.82s/it]
 55%|█████▌    | 8099/14725 [4:08:18<3:20:55,  1.82s/it]
 55%|█████▌    | 8100/14725 [4:08:20<3:23:05,  1.84s/it]
                                                        

 55%|█████▌    | 8100/14725 [4:08:20<3:23:05,  1.84s/it]
 55%|█████▌    | 8101/14725 [4:08:22<3:22:47,  1.84s/it]
 55%|█████▌    | 8102/14725 [4:08:24<3:23:09,  1.84s/it]
 55%|█████▌    | 8103/14725 [4:08:26<3:21:23
+0: {'loss': 0.8342, 'grad_norm': 0.7126597407664195, 'learning_rate': 2e-05, 'epoch': 0.55}
+0: ,  1.82s/it]
 55%|█████▌    | 8104/14725 [4:08:27<3:19:40,  1.81s/it]
 55%|█████▌    | 8105/14725 [4:08:29<3:18:59,  1.80s/it]
 55%|█████▌    | 8106/14725 [4:08:31<3:17:56,  1.79s/it]
 55%|█████▌    | 8107/14725 [4:08:33<3:16:56,  1.79s/it]
 55%|█████▌    | 8108/14725 [4:08:35<3:22:02,  1.83s/it]
 55%|█████▌    | 8109/14725 [4:08:36<3:20:53,  1.82s/it]
 55%|█████▌    | 8110/14725 [4:08:38<3:21:00,  1.82s/it]
                                                        

 55%|█████▌    | 8110/14725 [4:08:38<3:21:00,  1.82s/it]
 55%|█████▌    | 8111/14725 [4:08:40<3:20:20,  1.82s/it]
 55%|█████▌    | 8112/14725 [4:08:42<3:28:37,  1.89s/it]
 55%|█████▌    | 8113/14725 [4:08:44<3:23:37,  1.85s/it]
 55%|█████▌    | 8114/14725 [4:08:46<3:25:23,  1.86s/it]
 55%|█████▌    | 8115/14725 [4:08:48<3:23:28,  1.85s/it]
 55%|█████▌    | 8116/14725 [4:08:49<3:22:26
+0: {'loss': 0.8088, 'grad_norm': 0.7064967143187237, 'learning_rate': 2e-05, 'epoch': 0.55}
+0: ,  1.84s/it]
 55%|█████▌    | 8117/14725 [4:08:51<3:22:10,  1.84s/it]
 55%|█████▌    | 8118/14725 [4:08:53<3:21:19,  1.83s/it]
 55%|█████▌    | 8119/14725 [4:08:55<3:20:57,  1.83s/it]
 55%|█████▌    | 8120/14725 [4:08:57<3:19:58,  1.82s/it]
                                                        

 55%|█████▌    | 8120/14725 [4:08:57<3:19:58,  1.82s/it]
 55%|█████▌    | 8121/14725 [4:08:58<3:19:49,  1.82s/it]
 55%|█████▌    | 8122/14725 [4:09:00<3:19:19,  1.81s/it]
 55%|█████▌    | 8123/14725 [4:09:02<3:20:08,  1.82s/it]
 55%|█████▌    | 8124/14725 [4:09:04<3:18:06,  1.80s/it]
 55%|█████▌    | 8125/14725 [4:09:06<3:19:59,  1.82s/it]
 55%|█████▌    | 8126/14725 [4:09:08<3:20:08,  1.82s/it]
 55%|█████▌    | 8127/14725 [4:09:09<3:19:20,  1.81s/it]
 55%|█████▌    | 8128/14725 [4:09:11<3:18:00,  1.80s/it]
 55%|█████▌    | 8129/14725 [4:09:13<3:17:58
+0: {'loss': 0.8175, 'grad_norm': 0.7367817342138328, 'learning_rate': 2e-05, 'epoch': 0.55}
+0: {'loss': 0.8298, 'grad_norm': 0.7431684627606148, 'learning_rate': 2e-05, 'epoch': 0.55}
+0: ,  1.80s/it]
 55%|█████▌    | 8130/14725 [4:09:15<3:16:19,  1.79s/it]
                                                        

 55%|█████▌    | 8130/14725 [4:09:15<3:16:19,  1.79s/it]
 55%|█████▌    | 8131/14725 [4:09:17<3:18:23,  1.81s/it]
 55%|█████▌    | 8132/14725 [4:09:18<3:18:31,  1.81s/it]
 55%|█████▌    | 8133/14725 [4:09:20<3:19:17,  1.81s/it]
 55%|█████▌    | 8134/14725 [4:09:22<3:22:25,  1.84s/it]
 55%|█████▌    | 8135/14725 [4:09:24<3:19:47,  1.82s/it]
 55%|█████▌    | 8136/14725 [4:09:26<3:22:34,  1.84s/it]
 55%|█████▌    | 8137/14725 [4:09:28<3:20:11,  1.82s/it]
 55%|█████▌    | 8138/14725 [4:09:29<3:19:09,  1.81s/it]
 55%|█████▌    | 8139/14725 [4:09:31<3:20:31,  1.83s/it]
 55%|█████▌    | 8140/14725 [4:09:33<3:18:58,  1.81s/it]
                                                        

 55%|█████▌    | 8140/14725 [4:09:33<3:18:58,  1.81s/it
+0: {'loss': 0.8158, 'grad_norm': 0.7097407582595037, 'learning_rate': 2e-05, 'epoch': 0.55}
+0: ]
 55%|█████▌    | 8141/14725 [4:09:35<3:17:21,  1.80s/it]
 55%|█████▌    | 8142/14725 [4:09:37<3:16:54,  1.79s/it]
 55%|█████▌    | 8143/14725 [4:09:38<3:16:15,  1.79s/it]
 55%|█████▌    | 8144/14725 [4:09:40<3:16:04,  1.79s/it]
 55%|█████▌    | 8145/14725 [4:09:42<3:14:33,  1.77s/it]
 55%|█████▌    | 8146/14725 [4:09:44<3:14:07,  1.77s/it]
 55%|█████▌    | 8147/14725 [4:09:45<3:15:04,  1.78s/it]
 55%|█████▌    | 8148/14725 [4:09:47<3:22:19,  1.85s/it]
 55%|█████▌    | 8149/14725 [4:09:49<3:20:31,  1.83s/it]
 55%|█████▌    | 8150/14725 [4:09:51<3:20:10,  1.83s/it]
                                                        

 55%|█████▌    | 8150/14725 [4:09:51<3:20:10,  1.83s/it]
 55%|█████▌    | 8151/14725 [4:09:53<3:19:13,  1.82s/it]
 55%|█████▌    | 8152/14725 [4:09:55<3:18:41,  1.81s/it]
 55%|█████▌    | 8153/14725 [4:09:56<3:19:20,  1.82s/it
+0: {'loss': 0.804, 'grad_norm': 0.7478003300269807, 'learning_rate': 2e-05, 'epoch': 0.55}
+0: ]
 55%|█████▌    | 8154/14725 [4:09:58<3:17:38,  1.80s/it]
 55%|█████▌    | 8155/14725 [4:10:00<3:18:42,  1.81s/it]
 55%|█████▌    | 8156/14725 [4:10:02<3:17:00,  1.80s/it]
 55%|█████▌    | 8157/14725 [4:10:04<3:18:00,  1.81s/it]
 55%|█████▌    | 8158/14725 [4:10:05<3:16:31,  1.80s/it]
 55%|█████▌    | 8159/14725 [4:10:07<3:20:05,  1.83s/it]
 55%|█████▌    | 8160/14725 [4:10:09<3:23:52,  1.86s/it]
                                                        

 55%|█████▌    | 8160/14725 [4:10:09<3:23:52,  1.86s/it]
 55%|█████▌    | 8161/14725 [4:10:11<3:20:41,  1.83s/it]
 55%|█████▌    | 8162/14725 [4:10:13<3:20:24,  1.83s/it]
 55%|█████▌    | 8163/14725 [4:10:15<3:18:05,  1.81s/it]
 55%|█████▌    | 8164/14725 [4:10:16<3:16:54,  1.80s/it]
 55%|█████▌    | 8165/14725 [4:10:18<3:17:24,  1.81s/it]
 55%|█████▌    | 8166/14725 [4:10:20<3:18:41,  1.82s/it
+0: {'loss': 0.8225, 'grad_norm': 0.7105845772827319, 'learning_rate': 2e-05, 'epoch': 0.55}
+0: ]
 55%|█████▌    | 8167/14725 [4:10:22<3:17:43,  1.81s/it]
 55%|█████▌    | 8168/14725 [4:10:24<3:16:15,  1.80s/it]
 55%|█████▌    | 8169/14725 [4:10:25<3:15:34,  1.79s/it]
 55%|█████▌    | 8170/14725 [4:10:27<3:16:09,  1.80s/it]
                                                        

 55%|█████▌    | 8170/14725 [4:10:27<3:16:09,  1.80s/it]
 55%|█████▌    | 8171/14725 [4:10:29<3:16:50,  1.80s/it]
 55%|█████▌    | 8172/14725 [4:10:31<3:16:36,  1.80s/it]
 56%|█████▌    | 8173/14725 [4:10:33<3:15:37,  1.79s/it]
 56%|█████▌    | 8174/14725 [4:10:34<3:14:18,  1.78s/it]
 56%|█████▌    | 8175/14725 [4:10:36<3:13:31,  1.77s/it]
 56%|█████▌    | 8176/14725 [4:10:38<3:15:00,  1.79s/it]
 56%|█████▌    | 8177/14725 [4:10:40<3:16:03,  1.80s/it]
 56%|█████▌    | 8178/14725 [4:10:42<3:17:37,  1.81s/it]
 56%|█████▌    | 8179/14725 [4:10:43<3:17:40,  1.81s/it
+0: {'loss': 0.7996, 'grad_norm': 0.6858439326467098, 'learning_rate': 2e-05, 'epoch': 0.56}
+0: {'loss': 0.8123, 'grad_norm': 0.7075972364054008, 'learning_rate': 2e-05, 'epoch': 0.56}
+0: ]
 56%|█████▌    | 8180/14725 [4:10:45<3:16:29,  1.80s/it]
                                                        

 56%|█████▌    | 8180/14725 [4:10:45<3:16:29,  1.80s/it]
 56%|█████▌    | 8181/14725 [4:10:47<3:18:55,  1.82s/it]
 56%|█████▌    | 8182/14725 [4:10:49<3:17:14,  1.81s/it]
 56%|█████▌    | 8183/14725 [4:10:51<3:17:25,  1.81s/it]
 56%|█████▌    | 8184/14725 [4:10:52<3:19:27,  1.83s/it]
 56%|█████▌    | 8185/14725 [4:10:54<3:17:31,  1.81s/it]
 56%|█████▌    | 8186/14725 [4:10:56<3:16:25,  1.80s/it]
 56%|█████▌    | 8187/14725 [4:10:58<3:16:30,  1.80s/it]
 56%|█████▌    | 8188/14725 [4:11:00<3:17:55,  1.82s/it]
 56%|█████▌    | 8189/14725 [4:11:01<3:17:44,  1.82s/it]
 56%|█████▌    | 8190/14725 [4:11:03<3:18:04,  1.82s/it]
                                                        

 56%|█████▌    | 8190/14725 [4:11:03<3:18:04,  1.82s/it]
 56%|█�
+0: {'loss': 0.831, 'grad_norm': 0.7275397598103428, 'learning_rate': 2e-05, 'epoch': 0.56}
+0: ��███▌    | 8191/14725 [4:11:05<3:17:34,  1.81s/it]
 56%|█████▌    | 8192/14725 [4:11:07<3:15:42,  1.80s/it]
 56%|█████▌    | 8193/14725 [4:11:09<3:16:40,  1.81s/it]
 56%|█████▌    | 8194/14725 [4:11:11<3:16:47,  1.81s/it]
 56%|█████▌    | 8195/14725 [4:11:12<3:17:21,  1.81s/it]
 56%|█████▌    | 8196/14725 [4:11:14<3:16:45,  1.81s/it]
 56%|█████▌    | 8197/14725 [4:11:16<3:17:53,  1.82s/it]
 56%|█████▌    | 8198/14725 [4:11:18<3:17:34,  1.82s/it]
 56%|█████▌    | 8199/14725 [4:11:20<3:16:45,  1.81s/it]
 56%|█████▌    | 8200/14725 [4:11:21<3:18:40,  1.83s/it]
                                                        

 56%|█████▌    | 8200/14725 [4:11:21<3:18:40,  1.83s/it]
 56%|█████▌    | 8201/14725 [4:11:23<3:18:56,  1.83s/it]
 56%|█████▌    | 8202/14725 [4:11:25<3:16:51,  1.81s/it]
 56%|█████▌    | 8203/14725 [4:11:27<3:15:48,  1.80s/it]
 56%|█�
+0: {'loss': 0.8333, 'grad_norm': 0.7509075377804317, 'learning_rate': 2e-05, 'epoch': 0.56}
+0: ��███▌    | 8204/14725 [4:11:29<3:17:03,  1.81s/it]
 56%|█████▌    | 8205/14725 [4:11:30<3:15:32,  1.80s/it]
 56%|█████▌    | 8206/14725 [4:11:32<3:19:30,  1.84s/it]
 56%|█████▌    | 8207/14725 [4:11:34<3:18:47,  1.83s/it]
 56%|█████▌    | 8208/14725 [4:11:36<3:16:56,  1.81s/it]
 56%|█████▌    | 8209/14725 [4:11:38<3:16:04,  1.81s/it]
 56%|█████▌    | 8210/14725 [4:11:39<3:14:03,  1.79s/it]
                                                        

 56%|█████▌    | 8210/14725 [4:11:39<3:14:03,  1.79s/it]
 56%|█████▌    | 8211/14725 [4:11:41<3:14:54,  1.80s/it]
 56%|█████▌    | 8212/14725 [4:11:43<3:13:48,  1.79s/it]
 56%|█████▌    | 8213/14725 [4:11:45<3:12:45,  1.78s/it]
 56%|█████▌    | 8214/14725 [4:11:47<3:12:51,  1.78s/it]
 56%|█████▌    | 8215/14725 [4:11:48<3:15:34,  1.80s/it]
 56%|█████▌    | 8216/14725 [4:11:50<3:18:43,  1.83s/it]
 56%|█�
+0: {'loss': 0.8231, 'grad_norm': 0.6925781270057869, 'learning_rate': 2e-05, 'epoch': 0.56}
+0: ��███▌    | 8217/14725 [4:11:52<3:17:59,  1.83s/it]
 56%|█████▌    | 8218/14725 [4:11:54<3:17:36,  1.82s/it]
 56%|█████▌    | 8219/14725 [4:11:56<3:15:23,  1.80s/it]
 56%|█████▌    | 8220/14725 [4:11:57<3:13:50,  1.79s/it]
                                                        

 56%|█████▌    | 8220/14725 [4:11:57<3:13:50,  1.79s/it]
 56%|█████▌    | 8221/14725 [4:11:59<3:15:07,  1.80s/it]
 56%|█████▌    | 8222/14725 [4:12:01<3:16:28,  1.81s/it]
 56%|█████▌    | 8223/14725 [4:12:03<3:14:31,  1.80s/it]
 56%|█████▌    | 8224/14725 [4:12:05<3:13:33,  1.79s/it]
 56%|█████▌    | 8225/14725 [4:12:06<3:12:49,  1.78s/it]
 56%|█████▌    | 8226/14725 [4:12:08<3:14:34,  1.80s/it]
 56%|█████▌    | 8227/14725 [4:12:10<3:13:56,  1.79s/it]
 56%|█████▌    | 8228/14725 [4:12:12<3:13:46,  1.79s/it]
 56%|█████▌    | 8229/14725 [4:12:14<3:14:15,  1.79s/it]
 56%|█�
+0: {'loss': 0.8149, 'grad_norm': 0.7213906253142958, 'learning_rate': 2e-05, 'epoch': 0.56}
+0: {'loss': 0.8193, 'grad_norm': 0.7050129340965624, 'learning_rate': 2e-05, 'epoch': 0.56}
+0: ��███▌    | 8230/14725 [4:12:15<3:14:14,  1.79s/it]
                                                        

 56%|█████▌    | 8230/14725 [4:12:15<3:14:14,  1.79s/it]
 56%|█████▌    | 8231/14725 [4:12:17<3:15:27,  1.81s/it]
 56%|█████▌    | 8232/14725 [4:12:19<3:16:20,  1.81s/it]
 56%|█████▌    | 8233/14725 [4:12:21<3:16:00,  1.81s/it]
 56%|█████▌    | 8234/14725 [4:12:23<3:17:26,  1.83s/it]
 56%|█████▌    | 8235/14725 [4:12:25<3:15:32,  1.81s/it]
 56%|█████▌    | 8236/14725 [4:12:26<3:13:32,  1.79s/it]
 56%|█████▌    | 8237/14725 [4:12:28<3:15:34,  1.81s/it]
 56%|█████▌    | 8238/14725 [4:12:30<3:14:26,  1.80s/it]
 56%|█████▌    | 8239/14725 [4:12:32<3:14:59,  1.80s/it]
 56%|█████▌    | 8240/14725 [4:12:34<3:16:09,  1.81s/it]
                                                        

 56%|█████▌    | 8240/14725 [4:12:34<3:16:09,  1.81s/it]
 56%|█████
+0: {'loss': 0.8277, 'grad_norm': 0.7001681297476057, 'learning_rate': 2e-05, 'epoch': 0.56}
+0: ▌    | 8241/14725 [4:12:35<3:15:29,  1.81s/it]
 56%|█████▌    | 8242/14725 [4:12:37<3:16:47,  1.82s/it]
 56%|█████▌    | 8243/14725 [4:12:39<3:16:16,  1.82s/it]
 56%|█████▌    | 8244/14725 [4:12:41<3:17:33,  1.83s/it]
 56%|█████▌    | 8245/14725 [4:12:43<3:18:24,  1.84s/it]
 56%|█████▌    | 8246/14725 [4:12:45<3:16:36,  1.82s/it]
 56%|█████▌    | 8247/14725 [4:12:46<3:16:33,  1.82s/it]
 56%|█████▌    | 8248/14725 [4:12:48<3:15:26,  1.81s/it]
 56%|█████▌    | 8249/14725 [4:12:50<3:14:07,  1.80s/it]
 56%|█████▌    | 8250/14725 [4:12:52<3:15:06,  1.81s/it]
                                                        

 56%|█████▌    | 8250/14725 [4:12:52<3:15:06,  1.81s/it]
 56%|█████▌    | 8251/14725 [4:12:54<3:14:40,  1.80s/it]
 56%|█████▌    | 8252/14725 [4:12:55<3:16:16,  1.82s/it]
 56%|█████▌    | 8253/14725 [4:12:57<3:15:12,  1.81s/it]
 56%|█████
+0: {'loss': 0.8172, 'grad_norm': 0.7548823203300649, 'learning_rate': 2e-05, 'epoch': 0.56}
+0: ▌    | 8254/14725 [4:12:59<3:17:12,  1.83s/it]
 56%|█████▌    | 8255/14725 [4:13:01<3:16:18,  1.82s/it]
 56%|█████▌    | 8256/14725 [4:13:03<3:13:54,  1.80s/it]
 56%|█████▌    | 8257/14725 [4:13:04<3:12:36,  1.79s/it]
 56%|█████▌    | 8258/14725 [4:13:06<3:12:10,  1.78s/it]
 56%|█████▌    | 8259/14725 [4:13:08<3:13:22,  1.79s/it]
 56%|█████▌    | 8260/14725 [4:13:10<3:13:53,  1.80s/it]
                                                        

 56%|█████▌    | 8260/14725 [4:13:10<3:13:53,  1.80s/it]
 56%|█████▌    | 8261/14725 [4:13:12<3:13:43,  1.80s/it]
 56%|█████▌    | 8262/14725 [4:13:13<3:15:28,  1.81s/it]
 56%|█████▌    | 8263/14725 [4:13:15<3:13:52,  1.80s/it]
 56%|█████▌    | 8264/14725 [4:13:17<3:16:12,  1.82s/it]
 56%|█████▌    | 8265/14725 [4:13:19<3:14:22,  1.81s/it]
 56%|█████▌    | 8266/14725 [4:13:21<3:13:40,  1.80s/it]
 56%|█████
+0: {'loss': 0.8126, 'grad_norm': 0.7788985632367709, 'learning_rate': 2e-05, 'epoch': 0.56}
+0: ▌    | 8267/14725 [4:13:23<3:19:10,  1.85s/it]
 56%|█████▌    | 8268/14725 [4:13:24<3:19:01,  1.85s/it]
 56%|█████▌    | 8269/14725 [4:13:26<3:20:11,  1.86s/it]
 56%|█████▌    | 8270/14725 [4:13:28<3:19:59,  1.86s/it]
                                                        

 56%|█████▌    | 8270/14725 [4:13:28<3:19:59,  1.86s/it]
 56%|█████▌    | 8271/14725 [4:13:30<3:19:22,  1.85s/it]
 56%|█████▌    | 8272/14725 [4:13:32<3:17:53,  1.84s/it]
 56%|█████▌    | 8273/14725 [4:13:34<3:17:03,  1.83s/it]
 56%|█████▌    | 8274/14725 [4:13:35<3:14:59,  1.81s/it]
 56%|█████▌    | 8275/14725 [4:13:37<3:13:48,  1.80s/it]
 56%|█████▌    | 8276/14725 [4:13:39<3:13:54,  1.80s/it]
 56%|█████▌    | 8277/14725 [4:13:41<3:12:12,  1.79s/it]
 56%|█████▌    | 8278/14725 [4:13:43<3:12:06,  1.79s/it]
 56%|█████▌    | 8279/14725 [4:13:44<3:15:39,  1.82s/it]
 56%|█████
+0: {'loss': 0.8257, 'grad_norm': 0.7223189673206482, 'learning_rate': 2e-05, 'epoch': 0.56}
+0: {'loss': 0.8267, 'grad_norm': 0.7611271047387916, 'learning_rate': 2e-05, 'epoch': 0.56}
+0: ▌    | 8280/14725 [4:13:46<3:16:23,  1.83s/it]
                                                        

 56%|█████▌    | 8280/14725 [4:13:46<3:16:23,  1.83s/it]
 56%|█████▌    | 8281/14725 [4:13:48<3:15:22,  1.82s/it]
 56%|█████▌    | 8282/14725 [4:13:50<3:13:16,  1.80s/it]
 56%|█████▋    | 8283/14725 [4:13:52<3:12:10,  1.79s/it]
 56%|█████▋    | 8284/14725 [4:13:53<3:14:47,  1.81s/it]
 56%|█████▋    | 8285/14725 [4:13:55<3:13:33,  1.80s/it]
 56%|█████▋    | 8286/14725 [4:13:57<3:12:40,  1.80s/it]
 56%|█████▋    | 8287/14725 [4:13:59<3:13:26,  1.80s/it]
 56%|█████▋    | 8288/14725 [4:14:01<3:14:49,  1.82s/it]
 56%|█████▋    | 8289/14725 [4:14:03<3:17:13,  1.84s/it]
 56%|█████▋    | 8290/14725 [4:14:04<3:16:59,  1.84s/it]
                                                        

 56%|█████▋    | 8290/14725 [4:14:04<3:16:59,  1.84s/it]
 56%|█████▋    | 82
+0: {'loss': 0.804, 'grad_norm': 0.7427270293268072, 'learning_rate': 2e-05, 'epoch': 0.56}
+0: 91/14725 [4:14:06<3:15:15,  1.82s/it]
 56%|█████▋    | 8292/14725 [4:14:08<3:14:54,  1.82s/it]
 56%|█████▋    | 8293/14725 [4:14:10<3:15:38,  1.82s/it]
 56%|█████▋    | 8294/14725 [4:14:12<3:14:33,  1.82s/it]
 56%|█████▋    | 8295/14725 [4:14:13<3:13:41,  1.81s/it]
 56%|█████▋    | 8296/14725 [4:14:15<3:12:30,  1.80s/it]
 56%|█████▋    | 8297/14725 [4:14:17<3:13:34,  1.81s/it]
 56%|█████▋    | 8298/14725 [4:14:19<3:13:34,  1.81s/it]
 56%|█████▋    | 8299/14725 [4:14:21<3:15:56,  1.83s/it]
 56%|█████▋    | 8300/14725 [4:14:22<3:14:09,  1.81s/it]
                                                        

 56%|█████▋    | 8300/14725 [4:14:22<3:14:09,  1.81s/it]
 56%|█████▋    | 8301/14725 [4:14:24<3:13:15,  1.80s/it]
 56%|█████▋    | 8302/14725 [4:14:26<3:11:42,  1.79s/it]
 56%|█████▋    | 8303/14725 [4:14:28<3:11:29,  1.79s/it]
 56%|█████▋    | 83
+0: {'loss': 0.8188, 'grad_norm': 0.6619416717237581, 'learning_rate': 2e-05, 'epoch': 0.56}
+0: 04/14725 [4:14:30<3:11:44,  1.79s/it]
 56%|█████▋    | 8305/14725 [4:14:31<3:10:08,  1.78s/it]
 56%|█████▋    | 8306/14725 [4:14:33<3:10:57,  1.78s/it]
 56%|█████▋    | 8307/14725 [4:14:35<3:11:08,  1.79s/it]
 56%|█████▋    | 8308/14725 [4:14:37<3:12:01,  1.80s/it]
 56%|█████▋    | 8309/14725 [4:14:39<3:12:26,  1.80s/it]
 56%|█████▋    | 8310/14725 [4:14:40<3:11:05,  1.79s/it]
                                                        

 56%|█████▋    | 8310/14725 [4:14:40<3:11:05,  1.79s/it]
 56%|█████▋    | 8311/14725 [4:14:42<3:11:37,  1.79s/it]
 56%|█████▋    | 8312/14725 [4:14:44<3:11:58,  1.80s/it]
 56%|█████▋    | 8313/14725 [4:14:46<3:14:31,  1.82s/it]
 56%|█████▋    | 8314/14725 [4:14:48<3:13:25,  1.81s/it]
 56%|█████▋    | 8315/14725 [4:14:49<3:12:49,  1.80s/it]
 56%|█████▋    | 8316/14725 [4:14:51<3:11:57,  1.80s/it]
 56%|█████▋    | 83
+0: {'loss': 0.816, 'grad_norm': 0.6900959689616386, 'learning_rate': 2e-05, 'epoch': 0.57}
+0: 17/14725 [4:14:53<3:13:45,  1.81s/it]
 56%|█████▋    | 8318/14725 [4:14:55<3:14:30,  1.82s/it]
 56%|█████▋    | 8319/14725 [4:14:57<3:17:15,  1.85s/it]
 57%|█████▋    | 8320/14725 [4:14:59<3:16:07,  1.84s/it]
                                                        

 57%|█████▋    | 8320/14725 [4:14:59<3:16:07,  1.84s/it]
 57%|█████▋    | 8321/14725 [4:15:00<3:16:11,  1.84s/it]
 57%|█████▋    | 8322/14725 [4:15:02<3:16:05,  1.84s/it]
 57%|█████▋    | 8323/14725 [4:15:04<3:14:01,  1.82s/it]
 57%|█████▋    | 8324/14725 [4:15:06<3:14:55,  1.83s/it]
 57%|█████▋    | 8325/14725 [4:15:08<3:14:00,  1.82s/it]
 57%|█████▋    | 8326/14725 [4:15:10<3:17:59,  1.86s/it]
 57%|█████▋    | 8327/14725 [4:15:11<3:18:23,  1.86s/it]
 57%|█████▋    | 8328/14725 [4:15:13<3:15:33,  1.83s/it]
 57%|█████▋    | 8329/14725 [4:15:15<3:15:14,  1.83s/it]
 57%|█████▋    | 83
+0: {'loss': 0.8186, 'grad_norm': 0.7023095452871879, 'learning_rate': 2e-05, 'epoch': 0.57}
+0: {'loss': 0.8169, 'grad_norm': 0.7641901043919338, 'learning_rate': 2e-05, 'epoch': 0.57}
+0: 30/14725 [4:15:17<3:16:39,  1.85s/it]
                                                        

 57%|█████▋    | 8330/14725 [4:15:17<3:16:39,  1.85s/it]
 57%|█████▋    | 8331/14725 [4:15:19<3:14:56,  1.83s/it]
 57%|█████▋    | 8332/14725 [4:15:21<3:13:19,  1.81s/it]
 57%|█████▋    | 8333/14725 [4:15:22<3:11:52,  1.80s/it]
 57%|█████▋    | 8334/14725 [4:15:24<3:10:38,  1.79s/it]
 57%|█████▋    | 8335/14725 [4:15:26<3:08:59,  1.77s/it]
 57%|█████▋    | 8336/14725 [4:15:28<3:09:45,  1.78s/it]
 57%|█████▋    | 8337/14725 [4:15:29<3:12:41,  1.81s/it]
 57%|█████▋    | 8338/14725 [4:15:31<3:10:41,  1.79s/it]
 57%|█████▋    | 8339/14725 [4:15:33<3:13:18,  1.82s/it]
 57%|█████▋    | 8340/14725 [4:15:35<3:12:28,  1.81s/it]
                                                        

 57%|█████▋    | 8340/14725 [4:15:35<3:12:28,  1.81s/it]
 57%|█████▋    | 8341/14725 [4
+0: {'loss': 0.8224, 'grad_norm': 0.7070952948540496, 'learning_rate': 2e-05, 'epoch': 0.57}
+0: :15:37<3:12:36,  1.81s/it]
 57%|█████▋    | 8342/14725 [4:15:39<3:11:58,  1.80s/it]
 57%|█████▋    | 8343/14725 [4:15:40<3:13:53,  1.82s/it]
 57%|█████▋    | 8344/14725 [4:15:42<3:12:15,  1.81s/it]
 57%|█████▋    | 8345/14725 [4:15:44<3:11:10,  1.80s/it]
 57%|█████▋    | 8346/14725 [4:15:46<3:11:55,  1.81s/it]
 57%|█████▋    | 8347/14725 [4:15:48<3:11:17,  1.80s/it]
 57%|█████▋    | 8348/14725 [4:15:49<3:14:08,  1.83s/it]
 57%|█████▋    | 8349/14725 [4:15:51<3:13:27,  1.82s/it]
 57%|█████▋    | 8350/14725 [4:15:53<3:13:46,  1.82s/it]
                                                        

 57%|█████▋    | 8350/14725 [4:15:53<3:13:46,  1.82s/it]
 57%|█████▋    | 8351/14725 [4:15:55<3:13:04,  1.82s/it]
 57%|█████▋    | 8352/14725 [4:15:57<3:11:08,  1.80s/it]
 57%|█████▋    | 8353/14725 [4:15:58<3:09:52,  1.79s/it]
 57%|█████▋    | 8354/14725 [4
+0: {'loss': 0.8097, 'grad_norm': 0.7175965150049296, 'learning_rate': 2e-05, 'epoch': 0.57}
+0: :16:00<3:09:38,  1.79s/it]
 57%|█████▋    | 8355/14725 [4:16:02<3:11:12,  1.80s/it]
 57%|█████▋    | 8356/14725 [4:16:04<3:11:00,  1.80s/it]
 57%|█████▋    | 8357/14725 [4:16:06<3:09:51,  1.79s/it]
 57%|█████▋    | 8358/14725 [4:16:07<3:09:37,  1.79s/it]
 57%|█████▋    | 8359/14725 [4:16:09<3:08:36,  1.78s/it]
 57%|█████▋    | 8360/14725 [4:16:11<3:09:07,  1.78s/it]
                                                        

 57%|█████▋    | 8360/14725 [4:16:11<3:09:07,  1.78s/it]
 57%|█████▋    | 8361/14725 [4:16:13<3:08:29,  1.78s/it]
 57%|█████▋    | 8362/14725 [4:16:14<3:09:59,  1.79s/it]
 57%|█████▋    | 8363/14725 [4:16:16<3:10:10,  1.79s/it]
 57%|█████▋    | 8364/14725 [4:16:18<3:09:20,  1.79s/it]
 57%|█████▋    | 8365/14725 [4:16:20<3:11:34,  1.81s/it]
 57%|█████▋    | 8366/14725 [4:16:22<3:10:20,  1.80s/it]
 57%|█████▋    | 8367/14725 [4
+0: {'loss': 0.7989, 'grad_norm': 0.660438790055729, 'learning_rate': 2e-05, 'epoch': 0.57}
+0: :16:23<3:09:35,  1.79s/it]
 57%|█████▋    | 8368/14725 [4:16:25<3:09:57,  1.79s/it]
 57%|█████▋    | 8369/14725 [4:16:27<3:12:58,  1.82s/it]
 57%|█████▋    | 8370/14725 [4:16:29<3:12:30,  1.82s/it]
                                                        

 57%|█████▋    | 8370/14725 [4:16:29<3:12:30,  1.82s/it]
 57%|█████▋    | 8371/14725 [4:16:31<3:17:17,  1.86s/it]
 57%|█████▋    | 8372/14725 [4:16:33<3:15:52,  1.85s/it]
 57%|█████▋    | 8373/14725 [4:16:35<3:16:13,  1.85s/it]
 57%|█████▋    | 8374/14725 [4:16:36<3:14:10,  1.83s/it]
 57%|█████▋    | 8375/14725 [4:16:38<3:14:28,  1.84s/it]
 57%|█████▋    | 8376/14725 [4:16:40<3:12:47,  1.82s/it]
 57%|█████▋    | 8377/14725 [4:16:42<3:12:11,  1.82s/it]
 57%|█████▋    | 8378/14725 [4:16:44<3:12:38,  1.82s/it]
 57%|█████▋    | 8379/14725 [4:16:45<3:10:26,  1.80s/it]
 57%|█████▋    | 8380/14725 [4
+0: {'loss': 0.8237, 'grad_norm': 0.7140571524241706, 'learning_rate': 2e-05, 'epoch': 0.57}
+0: {'loss': 0.8204, 'grad_norm': 0.6911797341303866, 'learning_rate': 2e-05, 'epoch': 0.57}
+0: :16:47<3:08:40,  1.78s/it]
                                                        

 57%|█████▋    | 8380/14725 [4:16:47<3:08:40,  1.78s/it]
 57%|█████▋    | 8381/14725 [4:16:49<3:08:36,  1.78s/it]
 57%|█████▋    | 8382/14725 [4:16:51<3:10:19,  1.80s/it]
 57%|█████▋    | 8383/14725 [4:16:53<3:12:03,  1.82s/it]
 57%|█████▋    | 8384/14725 [4:16:54<3:10:13,  1.80s/it]
 57%|█████▋    | 8385/14725 [4:16:56<3:09:02,  1.79s/it]
 57%|█████▋    | 8386/14725 [4:16:58<3:10:54,  1.81s/it]
 57%|█████▋    | 8387/14725 [4:17:00<3:13:13,  1.83s/it]
 57%|█████▋    | 8388/14725 [4:17:02<3:12:04,  1.82s/it]
 57%|█████▋    | 8389/14725 [4:17:04<3:12:51,  1.83s/it]
 57%|█████▋    | 8390/14725 [4:17:05<3:13:34,  1.83s/it]
                                                        

 57%|█████▋    | 8390/14725 [4:17:05<3:13:34,  1.83s/it]
 57%|█████▋    | 8391/14725 [4:17:07<3:11
+0: {'loss': 0.8142, 'grad_norm': 0.7125531218768577, 'learning_rate': 2e-05, 'epoch': 0.57}
+0: :03,  1.81s/it]
 57%|█████▋    | 8392/14725 [4:17:09<3:11:26,  1.81s/it]
 57%|█████▋    | 8393/14725 [4:17:11<3:12:21,  1.82s/it]
 57%|█████▋    | 8394/14725 [4:17:13<3:10:20,  1.80s/it]
 57%|█████▋    | 8395/14725 [4:17:14<3:10:59,  1.81s/it]
 57%|█████▋    | 8396/14725 [4:17:16<3:12:49,  1.83s/it]
 57%|█████▋    | 8397/14725 [4:17:18<3:15:08,  1.85s/it]
 57%|█████▋    | 8398/14725 [4:17:20<3:13:20,  1.83s/it]
 57%|█████▋    | 8399/14725 [4:17:22<3:12:44,  1.83s/it]
 57%|█████▋    | 8400/14725 [4:17:24<3:11:22,  1.82s/it]
                                                        

 57%|█████▋    | 8400/14725 [4:17:24<3:11:22,  1.82s/it]
 57%|█████▋    | 8401/14725 [4:17:25<3:12:14,  1.82s/it]
 57%|█████▋    | 8402/14725 [4:17:27<3:11:41,  1.82s/it]
 57%|█████▋    | 8403/14725 [4:17:29<3:10:42,  1.81s/it]
 57%|█████▋    | 8404/14725 [4:17:31<3:13
+0: {'loss': 0.8045, 'grad_norm': 0.7419015990179122, 'learning_rate': 2e-05, 'epoch': 0.57}
+0: :36,  1.84s/it]
 57%|█████▋    | 8405/14725 [4:17:33<3:13:28,  1.84s/it]
 57%|█████▋    | 8406/14725 [4:17:34<3:10:41,  1.81s/it]
 57%|█████▋    | 8407/14725 [4:17:36<3:10:38,  1.81s/it]
 57%|█████▋    | 8408/14725 [4:17:38<3:10:44,  1.81s/it]
 57%|█████▋    | 8409/14725 [4:17:40<3:09:56,  1.80s/it]
 57%|█████▋    | 8410/14725 [4:17:42<3:10:03,  1.81s/it]
                                                        

 57%|█████▋    | 8410/14725 [4:17:42<3:10:03,  1.81s/it]
 57%|█████▋    | 8411/14725 [4:17:43<3:09:54,  1.80s/it]
 57%|█████▋    | 8412/14725 [4:17:45<3:09:03,  1.80s/it]
 57%|█████▋    | 8413/14725 [4:17:47<3:07:55,  1.79s/it]
 57%|█████▋    | 8414/14725 [4:17:49<3:10:02,  1.81s/it]
 57%|█████▋    | 8415/14725 [4:17:51<3:11:11,  1.82s/it]
 57%|█████▋    | 8416/14725 [4:17:53<3:09:51,  1.81s/it]
 57%|█████▋    | 8417/14725 [4:17:54<3:08
+0: {'loss': 0.8067, 'grad_norm': 0.6943495181127969, 'learning_rate': 2e-05, 'epoch': 0.57}
+0: :07,  1.79s/it]
 57%|█████▋    | 8418/14725 [4:17:56<3:08:39,  1.79s/it]
 57%|█████▋    | 8419/14725 [4:17:58<3:10:31,  1.81s/it]
 57%|█████▋    | 8420/14725 [4:18:00<3:09:10,  1.80s/it]
                                                        

 57%|█████▋    | 8420/14725 [4:18:00<3:09:10,  1.80s/it]
 57%|█████▋    | 8421/14725 [4:18:01<3:07:34,  1.79s/it]
 57%|█████▋    | 8422/14725 [4:18:03<3:16:04,  1.87s/it]
 57%|█████▋    | 8423/14725 [4:18:05<3:15:51,  1.86s/it]
 57%|█████▋    | 8424/14725 [4:18:07<3:14:09,  1.85s/it]
 57%|█████▋    | 8425/14725 [4:18:09<3:14:11,  1.85s/it]
 57%|█████▋    | 8426/14725 [4:18:11<3:15:10,  1.86s/it]
 57%|█████▋    | 8427/14725 [4:18:13<3:13:46,  1.85s/it]
 57%|█████▋    | 8428/14725 [4:18:15<3:12:57,  1.84s/it]
 57%|█████▋    | 8429/14725 [4:18:16<3:12:20,  1.83s/it]
 57%|█████▋    | 8430/14725 [4:18:18<3:12
+0: {'loss': 0.817, 'grad_norm': 0.6594789001245707, 'learning_rate': 2e-05, 'epoch': 0.57}
+0: {'loss': 0.794, 'grad_norm': 0.7079718249779084, 'learning_rate': 2e-05, 'epoch': 0.57}
+0: :19,  1.83s/it]
                                                        

 57%|█████▋    | 8430/14725 [4:18:18<3:12:19,  1.83s/it]
 57%|█████▋    | 8431/14725 [4:18:20<3:10:37,  1.82s/it]
 57%|█████▋    | 8432/14725 [4:18:22<3:10:26,  1.82s/it]
 57%|█████▋    | 8433/14725 [4:18:24<3:13:16,  1.84s/it]
 57%|█████▋    | 8434/14725 [4:18:26<3:13:24,  1.84s/it]
 57%|█████▋    | 8435/14725 [4:18:27<3:12:21,  1.83s/it]
 57%|█████▋    | 8436/14725 [4:18:29<3:11:57,  1.83s/it]
 57%|█████▋    | 8437/14725 [4:18:31<3:10:16,  1.82s/it]
 57%|█████▋    | 8438/14725 [4:18:33<3:13:08,  1.84s/it]
 57%|█████▋    | 8439/14725 [4:18:35<3:11:22,  1.83s/it]
 57%|█████▋    | 8440/14725 [4:18:36<3:09:32,  1.81s/it]
                                                        

 57%|█████▋    | 8440/14725 [4:18:36<3:09:32,  1.81s/it]
 57%|█████▋    | 8441/14725 [4:18:38<3:09:52,  1.81s
+0: {'loss': 0.8171, 'grad_norm': 0.7153269844576379, 'learning_rate': 2e-05, 'epoch': 0.57}
+0: /it]
 57%|█████▋    | 8442/14725 [4:18:40<3:09:18,  1.81s/it]
 57%|█████▋    | 8443/14725 [4:18:42<3:08:01,  1.80s/it]
 57%|█████▋    | 8444/14725 [4:18:44<3:08:19,  1.80s/it]
 57%|█████▋    | 8445/14725 [4:18:45<3:07:24,  1.79s/it]
 57%|█████▋    | 8446/14725 [4:18:47<3:08:05,  1.80s/it]
 57%|█████▋    | 8447/14725 [4:18:49<3:11:04,  1.83s/it]
 57%|█████▋    | 8448/14725 [4:18:51<3:09:39,  1.81s/it]
 57%|█████▋    | 8449/14725 [4:18:53<3:09:19,  1.81s/it]
 57%|█████▋    | 8450/14725 [4:18:54<3:07:45,  1.80s/it]
                                                        

 57%|█████▋    | 8450/14725 [4:18:54<3:07:45,  1.80s/it]
 57%|█████▋    | 8451/14725 [4:18:56<3:07:22,  1.79s/it]
 57%|█████▋    | 8452/14725 [4:18:58<3:09:19,  1.81s/it]
 57%|█████▋    | 8453/14725 [4:19:00<3:09:33,  1.81s/it]
 57%|█████▋    | 8454/14725 [4:19:02<3:08:56,  1.81s
+0: {'loss': 0.8162, 'grad_norm': 0.6643906329877813, 'learning_rate': 2e-05, 'epoch': 0.57}
+0: /it]
 57%|█████▋    | 8455/14725 [4:19:03<3:07:42,  1.80s/it]
 57%|█████▋    | 8456/14725 [4:19:05<3:08:33,  1.80s/it]
 57%|█████▋    | 8457/14725 [4:19:07<3:08:16,  1.80s/it]
 57%|█████▋    | 8458/14725 [4:19:09<3:10:08,  1.82s/it]
 57%|█████▋    | 8459/14725 [4:19:11<3:09:55,  1.82s/it]
 57%|█████▋    | 8460/14725 [4:19:12<3:07:28,  1.80s/it]
                                                        

 57%|█████▋    | 8460/14725 [4:19:12<3:07:28,  1.80s/it]
 57%|█████▋    | 8461/14725 [4:19:14<3:07:05,  1.79s/it]
 57%|█████▋    | 8462/14725 [4:19:16<3:07:17,  1.79s/it]
 57%|█████▋    | 8463/14725 [4:19:18<3:06:35,  1.79s/it]
 57%|█████▋    | 8464/14725 [4:19:20<3:07:14,  1.79s/it]
 57%|█████▋    | 8465/14725 [4:19:22<3:09:00,  1.81s/it]
 57%|█████▋    | 8466/14725 [4:19:23<3:07:12,  1.79s/it]
 58%|█████▊    | 8467/14725 [4:19:25<3:05:55,  1.78s
+0: {'loss': 0.8075, 'grad_norm': 0.6653728415437977, 'learning_rate': 2e-05, 'epoch': 0.58}
+0: /it]
 58%|█████▊    | 8468/14725 [4:19:27<3:07:49,  1.80s/it]
 58%|█████▊    | 8469/14725 [4:19:29<3:06:32,  1.79s/it]
 58%|█████▊    | 8470/14725 [4:19:30<3:05:50,  1.78s/it]
                                                        

 58%|█████▊    | 8470/14725 [4:19:30<3:05:50,  1.78s/it]
 58%|█████▊    | 8471/14725 [4:19:32<3:05:10,  1.78s/it]
 58%|█████▊    | 8472/14725 [4:19:34<3:07:15,  1.80s/it]
 58%|█████▊    | 8473/14725 [4:19:36<3:08:41,  1.81s/it]
 58%|█████▊    | 8474/14725 [4:19:38<3:07:58,  1.80s/it]
 58%|█████▊    | 8475/14725 [4:19:39<3:06:43,  1.79s/it]
 58%|█████▊    | 8476/14725 [4:19:41<3:06:25,  1.79s/it]
 58%|█████▊    | 8477/14725 [4:19:43<3:06:27,  1.79s/it]
 58%|█████▊    | 8478/14725 [4:19:45<3:06:32,  1.79s/it]
 58%|█████▊    | 8479/14725 [4:19:47<3:07:38,  1.80s/it]
 58%|█████▊    | 8480/14725 [4:19:48<3:08:52,  1.81s
+0: {'loss': 0.8173, 'grad_norm': 0.6802591656788897, 'learning_rate': 2e-05, 'epoch': 0.58}
+0: {'loss': 0.8036, 'grad_norm': 0.7404815380239963, 'learning_rate': 2e-05, 'epoch': 0.58}
+0: /it]
                                                        

 58%|█████▊    | 8480/14725 [4:19:48<3:08:52,  1.81s/it]
 58%|█████▊    | 8481/14725 [4:19:50<3:08:34,  1.81s/it]
 58%|█████▊    | 8482/14725 [4:19:52<3:07:54,  1.81s/it]
 58%|█████▊    | 8483/14725 [4:19:54<3:06:26,  1.79s/it]
 58%|█████▊    | 8484/14725 [4:19:56<3:07:13,  1.80s/it]
 58%|█████▊    | 8485/14725 [4:19:57<3:05:47,  1.79s/it]
 58%|█████▊    | 8486/14725 [4:19:59<3:04:23,  1.77s/it]
 58%|█████▊    | 8487/14725 [4:20:01<3:06:11,  1.79s/it]
 58%|█████▊    | 8488/14725 [4:20:03<3:07:57,  1.81s/it]
 58%|█████▊    | 8489/14725 [4:20:05<3:06:42,  1.80s/it]
 58%|█████▊    | 8490/14725 [4:20:06<3:05:39,  1.79s/it]
                                                        

 58%|█████▊    | 8490/14725 [4:20:06<3:05:39,  1.79s/it]
 58%|█████▊    | 8491/14725 [4:20:08<3:07:23,  1.80s/it]
 58%|�
+0: {'loss': 0.8112, 'grad_norm': 0.8282545903650107, 'learning_rate': 2e-05, 'epoch': 0.58}
+0: ��████▊    | 8492/14725 [4:20:10<3:06:27,  1.79s/it]
 58%|█████▊    | 8493/14725 [4:20:12<3:06:30,  1.80s/it]
 58%|█████▊    | 8494/14725 [4:20:14<3:07:47,  1.81s/it]
 58%|█████▊    | 8495/14725 [4:20:15<3:07:21,  1.80s/it]
 58%|█████▊    | 8496/14725 [4:20:17<3:08:11,  1.81s/it]
 58%|█████▊    | 8497/14725 [4:20:19<3:06:01,  1.79s/it]
 58%|█████▊    | 8498/14725 [4:20:21<3:04:23,  1.78s/it]
 58%|█████▊    | 8499/14725 [4:20:22<3:04:28,  1.78s/it]
 58%|█████▊    | 8500/14725 [4:20:24<3:07:49,  1.81s/it]
                                                        

 58%|█████▊    | 8500/14725 [4:20:24<3:07:49,  1.81s/it]
 58%|█████▊    | 8501/14725 [4:20:26<3:08:26,  1.82s/it]
 58%|█████▊    | 8502/14725 [4:20:28<3:07:29,  1.81s/it]
 58%|█████▊    | 8503/14725 [4:20:30<3:07:55,  1.81s/it]
 58%|█████▊    | 8504/14725 [4:20:32<3:07:21,  1.81s/it]
 58%|�
+0: {'loss': 0.8125, 'grad_norm': 0.7087081286011331, 'learning_rate': 2e-05, 'epoch': 0.58}
+0: ��████▊    | 8505/14725 [4:20:33<3:07:37,  1.81s/it]
 58%|█████▊    | 8506/14725 [4:20:35<3:06:28,  1.80s/it]
 58%|█████▊    | 8507/14725 [4:20:37<3:07:00,  1.80s/it]
 58%|█████▊    | 8508/14725 [4:20:39<3:05:40,  1.79s/it]
 58%|█████▊    | 8509/14725 [4:20:41<3:06:27,  1.80s/it]
 58%|█████▊    | 8510/14725 [4:20:42<3:04:35,  1.78s/it]
                                                        

 58%|█████▊    | 8510/14725 [4:20:42<3:04:35,  1.78s/it]
 58%|█████▊    | 8511/14725 [4:20:44<3:04:58,  1.79s/it]
 58%|█████▊    | 8512/14725 [4:20:46<3:06:40,  1.80s/it]
 58%|█████▊    | 8513/14725 [4:20:48<3:08:35,  1.82s/it]
 58%|█████▊    | 8514/14725 [4:20:50<3:10:07,  1.84s/it]
 58%|█████▊    | 8515/14725 [4:20:52<3:12:01,  1.86s/it]
 58%|█████▊    | 8516/14725 [4:20:53<3:13:10,  1.87s/it]
 58%|█████▊    | 8517/14725 [4:20:55<3:12:29,  1.86s/it]
 58%|�
+0: {'loss': 0.8068, 'grad_norm': 0.6999210257608564, 'learning_rate': 2e-05, 'epoch': 0.58}
+0: ��████▊    | 8518/14725 [4:20:57<3:15:09,  1.89s/it]
 58%|█████▊    | 8519/14725 [4:20:59<3:11:47,  1.85s/it]
 58%|█████▊    | 8520/14725 [4:21:01<3:09:23,  1.83s/it]
                                                        

 58%|█████▊    | 8520/14725 [4:21:01<3:09:23,  1.83s/it]
 58%|█████▊    | 8521/14725 [4:21:03<3:10:46,  1.84s/it]
 58%|█████▊    | 8522/14725 [4:21:05<3:12:11,  1.86s/it]
 58%|█████▊    | 8523/14725 [4:21:07<3:15:02,  1.89s/it]
 58%|█████▊    | 8524/14725 [4:21:09<3:16:57,  1.91s/it]
 58%|█████▊    | 8525/14725 [4:21:10<3:14:11,  1.88s/it]
 58%|█████▊    | 8526/14725 [4:21:12<3:12:48,  1.87s/it]
 58%|█████▊    | 8527/14725 [4:21:14<3:11:25,  1.85s/it]
 58%|█████▊    | 8528/14725 [4:21:16<3:12:59,  1.87s/it]
 58%|█████▊    | 8529/14725 [4:21:18<3:09:36,  1.84s/it]
 58%|█████▊    | 8530/14725 [4:21:19<3:08:34,  1.83s/it]
      
+0: {'loss': 0.8094, 'grad_norm': 0.7864272509472994, 'learning_rate': 2e-05, 'epoch': 0.58}
+0: {'loss': 0.8315, 'grad_norm': 0.6888275046013002, 'learning_rate': 2e-05, 'epoch': 0.58}
+0:                                                   

 58%|█████▊    | 8530/14725 [4:21:19<3:08:34,  1.83s/it]
 58%|█████▊    | 8531/14725 [4:21:21<3:10:04,  1.84s/it]
 58%|█████▊    | 8532/14725 [4:21:23<3:09:23,  1.83s/it]
 58%|█████▊    | 8533/14725 [4:21:25<3:09:13,  1.83s/it]
 58%|█████▊    | 8534/14725 [4:21:27<3:09:44,  1.84s/it]
 58%|█████▊    | 8535/14725 [4:21:29<3:18:19,  1.92s/it]
 58%|█████▊    | 8536/14725 [4:21:31<3:15:34,  1.90s/it]
 58%|█████▊    | 8537/14725 [4:21:33<3:12:52,  1.87s/it]
 58%|█████▊    | 8538/14725 [4:21:35<3:19:40,  1.94s/it]
 58%|█████▊    | 8539/14725 [4:21:37<3:19:33,  1.94s/it]
 58%|█████▊    | 8540/14725 [4:21:38<3:15:02,  1.89s/it]
                                                        

 58%|█████▊    | 8540/14725 [4:21:38<3:15:02,  1.89s/it]
 58%|█████▊    | 8541/14725 [4:21:40<3:11:36,  1.86s/it]
 58%|████
+0: {'loss': 0.8036, 'grad_norm': 0.71356332811093, 'learning_rate': 2e-05, 'epoch': 0.58}
+0: █▊    | 8542/14725 [4:21:42<3:12:44,  1.87s/it]
 58%|█████▊    | 8543/14725 [4:21:44<3:09:54,  1.84s/it]
 58%|█████▊    | 8544/14725 [4:21:46<3:07:29,  1.82s/it]
 58%|█████▊    | 8545/14725 [4:21:47<3:06:12,  1.81s/it]
 58%|█████▊    | 8546/14725 [4:21:49<3:04:40,  1.79s/it]
 58%|█████▊    | 8547/14725 [4:21:51<3:04:14,  1.79s/it]
 58%|█████▊    | 8548/14725 [4:21:53<3:03:46,  1.79s/it]
 58%|█████▊    | 8549/14725 [4:21:54<3:03:07,  1.78s/it]
 58%|█████▊    | 8550/14725 [4:21:56<3:03:16,  1.78s/it]
                                                        

 58%|█████▊    | 8550/14725 [4:21:56<3:03:16,  1.78s/it]
 58%|█████▊    | 8551/14725 [4:21:58<3:03:12,  1.78s/it]
 58%|█████▊    | 8552/14725 [4:22:00<3:03:03,  1.78s/it]
 58%|█████▊    | 8553/14725 [4:22:02<3:02:06,  1.77s/it]
 58%|█████▊    | 8554/14725 [4:22:03<3:00:55,  1.76s/it]
 58%|████
+0: {'loss': 0.8057, 'grad_norm': 0.7250581035199921, 'learning_rate': 2e-05, 'epoch': 0.58}
+0: █▊    | 8555/14725 [4:22:05<3:01:15,  1.76s/it]
 58%|█████▊    | 8556/14725 [4:22:07<3:01:43,  1.77s/it]
 58%|█████▊    | 8557/14725 [4:22:09<3:01:18,  1.76s/it]
 58%|█████▊    | 8558/14725 [4:22:10<3:01:17,  1.76s/it]
 58%|█████▊    | 8559/14725 [4:22:12<3:00:35,  1.76s/it]
 58%|█████▊    | 8560/14725 [4:22:14<3:04:40,  1.80s/it]
                                                        

 58%|█████▊    | 8560/14725 [4:22:14<3:04:40,  1.80s/it]
 58%|█████▊    | 8561/14725 [4:22:16<3:05:15,  1.80s/it]
 58%|█████▊    | 8562/14725 [4:22:18<3:06:22,  1.81s/it]
 58%|█████▊    | 8563/14725 [4:22:19<3:05:15,  1.80s/it]
 58%|█████▊    | 8564/14725 [4:22:21<3:05:37,  1.81s/it]
 58%|█████▊    | 8565/14725 [4:22:23<3:05:20,  1.81s/it]
 58%|█████▊    | 8566/14725 [4:22:25<3:05:08,  1.80s/it]
 58%|█████▊    | 8567/14725 [4:22:27<3:04:08,  1.79s/it]
 58%|████
+0: {'loss': 0.7916, 'grad_norm': 0.6980396672063266, 'learning_rate': 2e-05, 'epoch': 0.58}
+0: █▊    | 8568/14725 [4:22:28<3:05:05,  1.80s/it]
 58%|█████▊    | 8569/14725 [4:22:30<3:04:56,  1.80s/it]
 58%|█████▊    | 8570/14725 [4:22:32<3:06:31,  1.82s/it]
                                                        

 58%|█████▊    | 8570/14725 [4:22:32<3:06:31,  1.82s/it]
 58%|█████▊    | 8571/14725 [4:22:34<3:07:20,  1.83s/it]
 58%|█████▊    | 8572/14725 [4:22:36<3:05:38,  1.81s/it]
 58%|█████▊    | 8573/14725 [4:22:38<3:06:32,  1.82s/it]
 58%|█████▊    | 8574/14725 [4:22:39<3:06:31,  1.82s/it]
 58%|█████▊    | 8575/14725 [4:22:41<3:04:02,  1.80s/it]
 58%|█████▊    | 8576/14725 [4:22:43<3:05:46,  1.81s/it]
 58%|█████▊    | 8577/14725 [4:22:45<3:04:21,  1.80s/it]
 58%|█████▊    | 8578/14725 [4:22:47<3:04:35,  1.80s/it]
 58%|█████▊    | 8579/14725 [4:22:48<3:08:16,  1.84s/it]
 58%|█████▊    | 8580/14725 [4:22:50<3:06:15,  1.82s/it]
                 
+0: {'loss': 0.8214, 'grad_norm': 0.7084436840702641, 'learning_rate': 2e-05, 'epoch': 0.58}
+0: {'loss': 0.7886, 'grad_norm': 0.7007290652327652, 'learning_rate': 2e-05, 'epoch': 0.58}
+0:                                        

 58%|█████▊    | 8580/14725 [4:22:50<3:06:15,  1.82s/it]
 58%|█████▊    | 8581/14725 [4:22:52<3:04:37,  1.80s/it]
 58%|█████▊    | 8582/14725 [4:22:54<3:02:47,  1.79s/it]
 58%|█████▊    | 8583/14725 [4:22:56<3:04:47,  1.81s/it]
 58%|█████▊    | 8584/14725 [4:22:57<3:05:18,  1.81s/it]
 58%|█████▊    | 8585/14725 [4:22:59<3:03:47,  1.80s/it]
 58%|█████▊    | 8586/14725 [4:23:01<3:02:57,  1.79s/it]
 58%|█████▊    | 8587/14725 [4:23:03<3:01:44,  1.78s/it]
 58%|█████▊    | 8588/14725 [4:23:05<3:02:23,  1.78s/it]
 58%|█████▊    | 8589/14725 [4:23:06<3:01:50,  1.78s/it]
 58%|█████▊    | 8590/14725 [4:23:08<3:01:54,  1.78s/it]
                                                        

 58%|█████▊    | 8590/14725 [4:23:08<3:01:54,  1.78s/it]
 58%|█████▊    | 8591/14725 [4:23:10<3:02:26,  1.78s/it]
 58%|█████▊    |
+0: {'loss': 0.8036, 'grad_norm': 0.7029506495841054, 'learning_rate': 2e-05, 'epoch': 0.58}
+0:  8592/14725 [4:23:12<3:03:40,  1.80s/it]
 58%|█████▊    | 8593/14725 [4:23:14<3:05:19,  1.81s/it]
 58%|█████▊    | 8594/14725 [4:23:15<3:03:41,  1.80s/it]
 58%|█████▊    | 8595/14725 [4:23:17<3:05:13,  1.81s/it]
 58%|█████▊    | 8596/14725 [4:23:19<3:07:19,  1.83s/it]
 58%|█████▊    | 8597/14725 [4:23:21<3:06:39,  1.83s/it]
 58%|█████▊    | 8598/14725 [4:23:23<3:05:38,  1.82s/it]
 58%|█████▊    | 8599/14725 [4:23:24<3:04:03,  1.80s/it]
 58%|█████▊    | 8600/14725 [4:23:26<3:01:54,  1.78s/it]
                                                        

 58%|█████▊    | 8600/14725 [4:23:26<3:01:54,  1.78s/it]
 58%|█████▊    | 8601/14725 [4:23:28<3:02:06,  1.78s/it]
 58%|█████▊    | 8602/14725 [4:23:30<3:00:54,  1.77s/it]
 58%|█████▊    | 8603/14725 [4:23:32<3:02:48,  1.79s/it]
 58%|█████▊    | 8604/14725 [4:23:33<3:02:06,  1.79s/it]
 58%|█████▊    |
+0: {'loss': 0.7891, 'grad_norm': 0.7772275243544029, 'learning_rate': 2e-05, 'epoch': 0.58}
+0:  8605/14725 [4:23:35<3:02:09,  1.79s/it]
 58%|█████▊    | 8606/14725 [4:23:37<3:02:43,  1.79s/it]
 58%|█████▊    | 8607/14725 [4:23:39<3:04:36,  1.81s/it]
 58%|█████▊    | 8608/14725 [4:23:40<3:02:47,  1.79s/it]
 58%|█████▊    | 8609/14725 [4:23:42<3:01:09,  1.78s/it]
 58%|█████▊    | 8610/14725 [4:23:44<3:02:28,  1.79s/it]
                                                        

 58%|█████▊    | 8610/14725 [4:23:44<3:02:28,  1.79s/it]
 58%|█████▊    | 8611/14725 [4:23:46<3:04:22,  1.81s/it]
 58%|█████▊    | 8612/14725 [4:23:48<3:08:44,  1.85s/it]
 58%|█████▊    | 8613/14725 [4:23:50<3:06:44,  1.83s/it]
 58%|█████▊    | 8614/14725 [4:23:51<3:06:50,  1.83s/it]
 59%|█████▊    | 8615/14725 [4:23:53<3:07:35,  1.84s/it]
 59%|█████▊    | 8616/14725 [4:23:55<3:06:58,  1.84s/it]
 59%|█████▊    | 8617/14725 [4:23:57<3:04:54,  1.82s/it]
 59%|█████▊    |
+0: {'loss': 0.8145, 'grad_norm': 0.7013482530065689, 'learning_rate': 2e-05, 'epoch': 0.59}
+0:  8618/14725 [4:23:59<3:03:59,  1.81s/it]
 59%|���████▊    | 8619/14725 [4:24:01<3:03:53,  1.81s/it]
 59%|█████▊    | 8620/14725 [4:24:02<3:06:44,  1.84s/it]
                                                        

 59%|█████▊    | 8620/14725 [4:24:02<3:06:44,  1.84s/it]
 59%|█████▊    | 8621/14725 [4:24:04<3:04:22,  1.81s/it]
 59%|█████▊    | 8622/14725 [4:24:06<3:04:19,  1.81s/it]
 59%|█████▊    | 8623/14725 [4:24:08<3:03:05,  1.80s/it]
 59%|█████▊    | 8624/14725 [4:24:10<3:02:00,  1.79s/it]
 59%|█████▊    | 8625/14725 [4:24:11<3:03:42,  1.81s/it]
 59%|█████▊    | 8626/14725 [4:24:13<3:01:36,  1.79s/it]
 59%|█████▊    | 8627/14725 [4:24:15<3:02:08,  1.79s/it]
 59%|█████▊    | 8628/14725 [4:24:17<3:01:56,  1.79s/it]
 59%|█████▊    | 8629/14725 [4:24:18<3:00:57,  1.78s/it]
 59%|█████▊    | 8630/14725 [4:24:20<3:04:03,  1.81s/it]
                            
+0: {'loss': 0.8149, 'grad_norm': 0.6966354878998579, 'learning_rate': 2e-05, 'epoch': 0.59}
+0: {'loss': 0.8085, 'grad_norm': 0.7217307551316017, 'learning_rate': 2e-05, 'epoch': 0.59}
+0:                             

 59%|█████▊    | 8630/14725 [4:24:20<3:04:03,  1.81s/it]
 59%|█████▊    | 8631/14725 [4:24:22<3:04:12,  1.81s/it]
 59%|█████▊    | 8632/14725 [4:24:24<3:03:27,  1.81s/it]
 59%|█████▊    | 8633/14725 [4:24:26<3:02:51,  1.80s/it]
 59%|█████▊    | 8634/14725 [4:24:28<3:03:34,  1.81s/it]
 59%|█████▊    | 8635/14725 [4:24:29<3:02:04,  1.79s/it]
 59%|█████▊    | 8636/14725 [4:24:31<3:00:56,  1.78s/it]
 59%|█████▊    | 8637/14725 [4:24:33<2:59:58,  1.77s/it]
 59%|█████▊    | 8638/14725 [4:24:35<3:00:38,  1.78s/it]
 59%|█████▊    | 8639/14725 [4:24:37<3:02:54,  1.80s/it]
 59%|█████▊    | 8640/14725 [4:24:38<3:04:11,  1.82s/it]
                                                        

 59%|█████▊    | 8640/14725 [4:24:38<3:04:11,  1.82s/it]
 59%|█████▊    | 8641/14725 [4:24:40<3:04:46,  1.82s/it]
 59%|█████▊    | 8642/14725
+0: {'loss': 0.7979, 'grad_norm': 0.716156412084853, 'learning_rate': 2e-05, 'epoch': 0.59}
+0:  [4:24:42<3:04:05,  1.82s/it]
 59%|█████▊    | 8643/14725 [4:24:44<3:03:25,  1.81s/it]
 59%|█████▊    | 8644/14725 [4:24:46<3:03:28,  1.81s/it]
 59%|█████▊    | 8645/14725 [4:24:47<3:02:50,  1.80s/it]
 59%|█████▊    | 8646/14725 [4:24:49<3:04:21,  1.82s/it]
 59%|█████▊    | 8647/14725 [4:24:51<3:03:27,  1.81s/it]
 59%|█████▊    | 8648/14725 [4:24:53<3:02:08,  1.80s/it]
 59%|█████▊    | 8649/14725 [4:24:55<3:02:26,  1.80s/it]
 59%|█████▊    | 8650/14725 [4:24:56<3:01:19,  1.79s/it]
                                                        

 59%|█████▊    | 8650/14725 [4:24:56<3:01:19,  1.79s/it]
 59%|█████▉    | 8651/14725 [4:24:58<3:00:12,  1.78s/it]
 59%|█████▉    | 8652/14725 [4:25:00<3:00:55,  1.79s/it]
 59%|█████▉    | 8653/14725 [4:25:02<3:00:24,  1.78s/it]
 59%|█████▉    | 8654/14725 [4:25:03<3:00:01,  1.78s/it]
 59%|█████▉    | 8655/14725
+0: {'loss': 0.8183, 'grad_norm': 0.73952808691763, 'learning_rate': 2e-05, 'epoch': 0.59}
+0:  [4:25:05<2:59:54,  1.78s/it]
 59%|█████▉    | 8656/14725 [4:25:07<2:59:09,  1.77s/it]
 59%|█████▉    | 8657/14725 [4:25:09<3:12:41,  1.91s/it]
 59%|█████▉    | 8658/14725 [4:25:11<3:08:15,  1.86s/it]
 59%|█████▉    | 8659/14725 [4:25:13<3:05:16,  1.83s/it]
 59%|█████▉    | 8660/14725 [4:25:15<3:05:10,  1.83s/it]
                                                        

 59%|█████▉    | 8660/14725 [4:25:15<3:05:10,  1.83s/it]
 59%|█████▉    | 8661/14725 [4:25:16<3:04:00,  1.82s/it]
 59%|█████▉    | 8662/14725 [4:25:18<3:02:54,  1.81s/it]
 59%|█████▉    | 8663/14725 [4:25:20<3:00:45,  1.79s/it]
 59%|█████▉    | 8664/14725 [4:25:22<3:00:13,  1.78s/it]
 59%|█████▉    | 8665/14725 [4:25:23<2:59:45,  1.78s/it]
 59%|█████▉    | 8666/14725 [4:25:25<2:59:18,  1.78s/it]
 59%|█████▉    | 8667/14725 [4:25:27<2:59:16,  1.78s/it]
 59%|█████▉    | 8668/14725
+0: {'loss': 0.7933, 'grad_norm': 0.6768554017047037, 'learning_rate': 2e-05, 'epoch': 0.59}
+0:  [4:25:29<3:00:45,  1.79s/it]
 59%|█████▉    | 8669/14725 [4:25:31<3:02:25,  1.81s/it]
 59%|█████▉    | 8670/14725 [4:25:32<3:01:32,  1.80s/it]
                                                        

 59%|█████▉    | 8670/14725 [4:25:32<3:01:32,  1.80s/it]
 59%|█████▉    | 8671/14725 [4:25:34<3:00:13,  1.79s/it]
 59%|█████▉    | 8672/14725 [4:25:36<3:04:03,  1.82s/it]
 59%|█████▉    | 8673/14725 [4:25:38<3:05:18,  1.84s/it]
 59%|█████▉    | 8674/14725 [4:25:40<3:06:23,  1.85s/it]
 59%|████���▉    | 8675/14725 [4:25:42<3:04:07,  1.83s/it]
 59%|█████▉    | 8676/14725 [4:25:43<3:01:49,  1.80s/it]
 59%|█████▉    | 8677/14725 [4:25:45<3:01:03,  1.80s/it]
 59%|█████▉    | 8678/14725 [4:25:47<3:11:20,  1.90s/it]
 59%|█████▉    | 8679/14725 [4:25:49<3:07:45,  1.86s/it]
 59%|█████▉    | 8680/14725 [4:25:51<3:05:37,  1.84s/it]
                                       
+0: {'loss': 0.8192, 'grad_norm': 0.7697817847930293, 'learning_rate': 2e-05, 'epoch': 0.59}
+0: {'loss': 0.8472, 'grad_norm': 0.7506471029678404, 'learning_rate': 2e-05, 'epoch': 0.59}
+0:                  

 59%|█████▉    | 8680/14725 [4:25:51<3:05:37,  1.84s/it]
 59%|█████▉    | 8681/14725 [4:25:53<3:03:51,  1.83s/it]
 59%|█████▉    | 8682/14725 [4:25:54<3:04:29,  1.83s/it]
 59%|█████▉    | 8683/14725 [4:25:56<3:05:02,  1.84s/it]
 59%|█████▉    | 8684/14725 [4:25:58<3:04:00,  1.83s/it]
 59%|█████▉    | 8685/14725 [4:26:00<3:05:43,  1.85s/it]
 59%|█████▉    | 8686/14725 [4:26:02<3:03:26,  1.82s/it]
 59%|█████▉    | 8687/14725 [4:26:04<3:02:39,  1.82s/it]
 59%|█████▉    | 8688/14725 [4:26:05<3:00:23,  1.79s/it]
 59%|█████▉    | 8689/14725 [4:26:07<3:04:01,  1.83s/it]
 59%|█████▉    | 8690/14725 [4:26:09<3:14:23,  1.93s/it]
                                                        

 59%|█████▉    | 8690/14725 [4:26:09<3:14:23,  1.93s/it]
 59%|█████▉    | 8691/14725 [4:26:11<3:11:42,  1.91s/it]
 59%|█████▉    | 8692/14725 [4:26:13<3
+0: {'loss': 0.8181, 'grad_norm': 0.7162851712420635, 'learning_rate': 2e-05, 'epoch': 0.59}
+0: :06:55,  1.86s/it]
 59%|█████▉    | 8693/14725 [4:26:15<3:04:19,  1.83s/it]
 59%|█████▉    | 8694/14725 [4:26:17<3:04:31,  1.84s/it]
 59%|█████▉    | 8695/14725 [4:26:19<3:05:26,  1.85s/it]
 59%|█████▉    | 8696/14725 [4:26:20<3:05:15,  1.84s/it]
 59%|█████▉    | 8697/14725 [4:26:22<3:04:24,  1.84s/it]
 59%|█████▉    | 8698/14725 [4:26:24<3:03:11,  1.82s/it]
 59%|█████▉    | 8699/14725 [4:26:26<3:03:25,  1.83s/it]
 59%|█████▉    | 8700/14725 [4:26:28<3:01:28,  1.81s/it]
                                                        

 59%|█████▉    | 8700/14725 [4:26:28<3:01:28,  1.81s/it]
 59%|█████▉    | 8701/14725 [4:26:29<3:01:39,  1.81s/it]
 59%|█████▉    | 8702/14725 [4:26:31<3:02:14,  1.82s/it]
 59%|█████▉    | 8703/14725 [4:26:33<3:02:05,  1.81s/it]
 59%|█████▉    | 8704/14725 [4:26:35<3:03:43,  1.83s/it]
 59%|█████▉    | 8705/14725 [4:26:37<3
+0: {'loss': 0.7962, 'grad_norm': 0.7241999274695067, 'learning_rate': 2e-05, 'epoch': 0.59}
+0: :04:53,  1.84s/it]
 59%|█████▉    | 8706/14725 [4:26:39<3:13:55,  1.93s/it]
 59%|█████▉    | 8707/14725 [4:26:41<3:09:04,  1.89s/it]
 59%|█████▉    | 8708/14725 [4:26:42<3:05:09,  1.85s/it]
 59%|█████▉    | 8709/14725 [4:26:44<3:03:18,  1.83s/it]
 59%|█████▉    | 8710/14725 [4:26:46<3:06:14,  1.86s/it]
                                                        

 59%|█████▉    | 8710/14725 [4:26:46<3:06:14,  1.86s/it]
 59%|█████▉    | 8711/14725 [4:26:48<3:03:49,  1.83s/it]
 59%|█████▉    | 8712/14725 [4:26:50<3:01:58,  1.82s/it]
 59%|█████▉    | 8713/14725 [4:26:52<3:04:45,  1.84s/it]
 59%|█████▉    | 8714/14725 [4:26:53<3:03:29,  1.83s/it]
 59%|█████▉    | 8715/14725 [4:26:55<3:01:31,  1.81s/it]
 59%|█████▉    | 8716/14725 [4:26:57<3:00:36,  1.80s/it]
 59%|█████▉    | 8717/14725 [4:26:59<2:59:45,  1.80s/it]
 59%|█████▉    | 8718/14725 [4:27:01<2
+0: {'loss': 0.8233, 'grad_norm': 0.716116997678952, 'learning_rate': 2e-05, 'epoch': 0.59}
+0: :59:10,  1.79s/it]
 59%|█████▉    | 8719/14725 [4:27:02<2:59:17,  1.79s/it]
 59%|█████▉    | 8720/14725 [4:27:04<2:59:32,  1.79s/it]
                                                        

 59%|█████▉    | 8720/14725 [4:27:04<2:59:32,  1.79s/it]
 59%|█████▉    | 8721/14725 [4:27:06<3:01:44,  1.82s/it]
 59%|█████▉    | 8722/14725 [4:27:08<3:02:26,  1.82s/it]
 59%|█████▉    | 8723/14725 [4:27:10<3:00:45,  1.81s/it]
 59%|█████▉    | 8724/14725 [4:27:11<2:59:14,  1.79s/it]
 59%|█████▉    | 8725/14725 [4:27:13<2:59:16,  1.79s/it]
 59%|█████▉    | 8726/14725 [4:27:15<2:57:48,  1.78s/it]
 59%|█████▉    | 8727/14725 [4:27:17<2:59:29,  1.80s/it]
 59%|█████▉    | 8728/14725 [4:27:19<2:59:31,  1.80s/it]
 59%|█████▉    | 8729/14725 [4:27:20<3:01:59,  1.82s/it]
 59%|█████▉    | 8730/14725 [4:27:22<2:59:49,  1.80s/it]
                                                  
+0: {'loss': 0.8059, 'grad_norm': 0.6599578415021599, 'learning_rate': 2e-05, 'epoch': 0.59}
+0: {'loss': 0.8334, 'grad_norm': 0.7293876914742112, 'learning_rate': 2e-05, 'epoch': 0.59}
+0:       

 59%|█████▉    | 8730/14725 [4:27:22<2:59:49,  1.80s/it]
 59%|█████▉    | 8731/14725 [4:27:24<3:04:32,  1.85s/it]
 59%|█████▉    | 8732/14725 [4:27:26<3:02:09,  1.82s/it]
 59%|█████▉    | 8733/14725 [4:27:28<3:01:46,  1.82s/it]
 59%|█████▉    | 8734/14725 [4:27:29<3:01:11,  1.81s/it]
 59%|█████▉    | 8735/14725 [4:27:31<2:59:07,  1.79s/it]
 59%|█████▉    | 8736/14725 [4:27:33<2:59:15,  1.80s/it]
 59%|█████▉    | 8737/14725 [4:27:35<2:59:21,  1.80s/it]
 59%|█████▉    | 8738/14725 [4:27:37<3:00:19,  1.81s/it]
 59%|█████▉    | 8739/14725 [4:27:39<3:01:37,  1.82s/it]
 59%|█████▉    | 8740/14725 [4:27:40<3:01:21,  1.82s/it]
                                                        

 59%|█████▉    | 8740/14725 [4:27:40<3:01:21,  1.82s/it]
 59%|█████▉    | 8741/14725 [4:27:42<3:01:48,  1.82s/it]
 59%|█████▉    | 8742/14725 [4:27:44<2:59:41,  1.
+0: {'loss': 0.7954, 'grad_norm': 0.6712789481216833, 'learning_rate': 2e-05, 'epoch': 0.59}
+0: 80s/it]
 59%|█████▉    | 8743/14725 [4:27:46<2:59:24,  1.80s/it]
 59%|█████▉    | 8744/14725 [4:27:47<2:58:10,  1.79s/it]
 59%|█████▉    | 8745/14725 [4:27:49<2:57:50,  1.78s/it]
 59%|█████▉    | 8746/14725 [4:27:51<3:02:24,  1.83s/it]
 59%|█████▉    | 8747/14725 [4:27:53<3:00:31,  1.81s/it]
 59%|█████▉    | 8748/14725 [4:27:55<2:58:34,  1.79s/it]
 59%|█████▉    | 8749/14725 [4:27:56<2:58:53,  1.80s/it]
 59%|█████▉    | 8750/14725 [4:27:58<2:57:51,  1.79s/it]
                                                        

 59%|█████▉    | 8750/14725 [4:27:58<2:57:51,  1.79s/it]
 59%|█████▉    | 8751/14725 [4:28:00<2:57:03,  1.78s/it]
 59%|█████▉    | 8752/14725 [4:28:02<3:00:20,  1.81s/it]
 59%|█████▉    | 8753/14725 [4:28:04<2:59:12,  1.80s/it]
 59%|█████▉    | 8754/14725 [4:28:06<2:59:45,  1.81s/it]
 59%|█████▉    | 8755/14725 [4:28:07<2:58:04,  1.
+0: {'loss': 0.8098, 'grad_norm': 0.680969816539819, 'learning_rate': 2e-05, 'epoch': 0.59}
+0: 79s/it]
 59%|█████▉    | 8756/14725 [4:28:09<2:58:46,  1.80s/it]
 59%|█████▉    | 8757/14725 [4:28:11<2:58:35,  1.80s/it]
 59%|█████▉    | 8758/14725 [4:28:13<2:58:33,  1.80s/it]
 59%|█████▉    | 8759/14725 [4:28:14<2:58:34,  1.80s/it]
 59%|█████▉    | 8760/14725 [4:28:16<2:59:59,  1.81s/it]
                                                        

 59%|█████▉    | 8760/14725 [4:28:16<2:59:59,  1.81s/it]
 59%|█████▉    | 8761/14725 [4:28:18<2:58:41,  1.80s/it]
 60%|█████▉    | 8762/14725 [4:28:20<2:58:37,  1.80s/it]
 60%|█████▉    | 8763/14725 [4:28:22<2:59:06,  1.80s/it]
 60%|█████▉    | 8764/14725 [4:28:23<2:58:33,  1.80s/it]
 60%|█████▉    | 8765/14725 [4:28:25<2:58:01,  1.79s/it]
 60%|█████▉    | 8766/14725 [4:28:27<2:59:21,  1.81s/it]
 60%|█████▉    | 8767/14725 [4:28:29<2:59:04,  1.80s/it]
 60%|█████▉    | 8768/14725 [4:28:31<2:57:24,  1.
+0: {'loss': 0.7924, 'grad_norm': 0.6956380501030623, 'learning_rate': 2e-05, 'epoch': 0.6}
+0: {'loss': 0.7998, 'grad_norm': 0.728699249889858, 'learning_rate': 2e-05, 'epoch': 0.6}
+0: 79s/it]
 60%|█████▉    | 8769/14725 [4:28:32<2:57:26,  1.79s/it]
 60%|█████▉    | 8770/14725 [4:28:34<3:00:53,  1.82s/it]
                                                        

 60%|█████▉    | 8770/14725 [4:28:34<3:00:53,  1.82s/it]
 60%|█████▉    | 8771/14725 [4:28:36<3:00:16,  1.82s/it]
 60%|█████▉    | 8772/14725 [4:28:38<2:59:39,  1.81s/it]
 60%|█████▉    | 8773/14725 [4:28:40<2:59:14,  1.81s/it]
 60%|█████▉    | 8774/14725 [4:28:42<3:00:52,  1.82s/it]
 60%|█████▉    | 8775/14725 [4:28:43<2:59:16,  1.81s/it]
 60%|█████▉    | 8776/14725 [4:28:45<2:58:27,  1.80s/it]
 60%|█████▉    | 8777/14725 [4:28:47<2:59:54,  1.81s/it]
 60%|█████▉    | 8778/14725 [4:28:49<3:00:21,  1.82s/it]
 60%|█████▉    | 8779/14725 [4:28:51<2:58:41,  1.80s/it]
 60%|█████▉    | 8780/14725 [4:28:52<2:58:11,  1.80s/it]
                                                        

 60
+0: {'loss': 0.8305, 'grad_norm': 0.6875313576050754, 'learning_rate': 2e-05, 'epoch': 0.6}
+0: %|█████▉    | 8780/14725 [4:28:52<2:58:11,  1.80s/it]
 60%|█████▉    | 8781/14725 [4:28:54<2:58:36,  1.80s/it]
 60%|█████▉    | 8782/14725 [4:28:56<2:57:30,  1.79s/it]
 60%|█████▉    | 8783/14725 [4:28:58<3:00:53,  1.83s/it]
 60%|█████▉    | 8784/14725 [4:29:00<2:58:59,  1.81s/it]
 60%|█████▉    | 8785/14725 [4:29:01<2:58:46,  1.81s/it]
 60%|█████▉    | 8786/14725 [4:29:03<2:57:15,  1.79s/it]
 60%|█████▉    | 8787/14725 [4:29:05<2:57:47,  1.80s/it]
 60%|█████▉    | 8788/14725 [4:29:07<2:58:24,  1.80s/it]
 60%|█████▉    | 8789/14725 [4:29:09<2:58:41,  1.81s/it]
 60%|█████▉    | 8790/14725 [4:29:10<2:59:38,  1.82s/it]
                                                        

 60%|█████▉    | 8790/14725 [4:29:10<2:59:38,  1.82s/it]
 60%|█████▉    | 8791/14725 [4:29:12<2:58:27,  1.80s/it]
 60%|█████▉    | 8792/14725 [4:29:14<3:00:03,  1.82s/it]
 60
+0: {'loss': 0.796, 'grad_norm': 0.7295743270408935, 'learning_rate': 2e-05, 'epoch': 0.6}
+0: %|█████▉    | 8793/14725 [4:29:16<2:58:46,  1.81s/it]
 60%|█████▉    | 8794/14725 [4:29:18<2:58:19,  1.80s/it]
 60%|█████▉    | 8795/14725 [4:29:19<2:58:48,  1.81s/it]
 60%|█████▉    | 8796/14725 [4:29:21<2:56:47,  1.79s/it]
 60%|█████▉    | 8797/14725 [4:29:23<2:56:48,  1.79s/it]
 60%|█████▉    | 8798/14725 [4:29:25<2:56:26,  1.79s/it]
 60%|█████▉    | 8799/14725 [4:29:27<2:59:14,  1.81s/it]
 60%|█████▉    | 8800/14725 [4:29:29<3:00:02,  1.82s/it]
                                                        

 60%|█████▉    | 8800/14725 [4:29:29<3:00:02,  1.82s/it]
 60%|█████▉    | 8801/14725 [4:29:30<3:00:01,  1.82s/it]
 60%|█████▉    | 8802/14725 [4:29:32<3:00:53,  1.83s/it]
 60%|█████▉    | 8803/14725 [4:29:34<2:59:17,  1.82s/it]
 60%|█████▉    | 8804/14725 [4:29:36<2:58:08,  1.81s/it]
 60%|█████▉    | 8805/14725 [4:29:38<2:56:23,  1.79s/it]
 60
+0: {'loss': 0.8068, 'grad_norm': 0.7050552523550772, 'learning_rate': 2e-05, 'epoch': 0.6}
+0: %|█████▉    | 8806/14725 [4:29:39<2:58:31,  1.81s/it]
 60%|█████▉    | 8807/14725 [4:29:41<2:58:57,  1.81s/it]
 60%|█████▉    | 8808/14725 [4:29:43<2:56:54,  1.79s/it]
 60%|█████▉    | 8809/14725 [4:29:45<2:58:34,  1.81s/it]
 60%|█████▉    | 8810/14725 [4:29:47<3:02:14,  1.85s/it]
                                                        

 60%|█████▉    | 8810/14725 [4:29:47<3:02:14,  1.85s/it]
 60%|█████▉    | 8811/14725 [4:29:49<3:00:26,  1.83s/it]
 60%|█████▉    | 8812/14725 [4:29:50<2:58:50,  1.81s/it]
 60%|█████▉    | 8813/14725 [4:29:52<2:59:57,  1.83s/it]
 60%|█████▉    | 8814/14725 [4:29:54<3:00:09,  1.83s/it]
 60%|█████▉    | 8815/14725 [4:29:56<2:58:58,  1.82s/it]
 60%|█████▉    | 8816/14725 [4:29:58<2:58:00,  1.81s/it]
 60%|█████▉    | 8817/14725 [4:29:59<2:58:24,  1.81s/it]
 60%|█████▉    | 8818/14725 [4:30:01<2:57:22,  1.80s/it]
 60
+0: {'loss': 0.7938, 'grad_norm': 0.7068828899740219, 'learning_rate': 2e-05, 'epoch': 0.6}
+0: {'loss': 0.8197, 'grad_norm': 0.7029598040768686, 'learning_rate': 2e-05, 'epoch': 0.6}
+0: %|█████▉    | 8819/14725 [4:30:03<2:58:41,  1.82s/it]
 60%|█████▉    | 8820/14725 [4:30:05<2:58:51,  1.82s/it]
                                                        

 60%|█████▉    | 8820/14725 [4:30:05<2:58:51,  1.82s/it]
 60%|█████▉    | 8821/14725 [4:30:07<2:59:18,  1.82s/it]
 60%|█████▉    | 8822/14725 [4:30:09<3:00:09,  1.83s/it]
 60%|█████▉    | 8823/14725 [4:30:10<2:58:39,  1.82s/it]
 60%|█████▉    | 8824/14725 [4:30:12<2:58:24,  1.81s/it]
 60%|█████▉    | 8825/14725 [4:30:14<2:59:01,  1.82s/it]
 60%|█████▉    | 8826/14725 [4:30:16<2:57:33,  1.81s/it]
 60%|█████▉    | 8827/14725 [4:30:17<2:57:05,  1.80s/it]
 60%|█████▉    | 8828/14725 [4:30:19<2:57:42,  1.81s/it]
 60%|█████▉    | 8829/14725 [4:30:21<3:00:11,  1.83s/it]
 60%|█████▉    | 8830/14725 [4:30:23<2:59:55,  1.83s/it]
                                                        

 60%|███
+0: {'loss': 0.81, 'grad_norm': 0.6817538940077794, 'learning_rate': 2e-05, 'epoch': 0.6}
+0: ██▉    | 8830/14725 [4:30:23<2:59:55,  1.83s/it]
 60%|█████▉    | 8831/14725 [4:30:25<3:02:06,  1.85s/it]
 60%|█████▉    | 8832/14725 [4:30:27<2:59:38,  1.83s/it]
 60%|█████▉    | 8833/14725 [4:30:28<2:57:17,  1.81s/it]
 60%|█████▉    | 8834/14725 [4:30:30<2:58:45,  1.82s/it]
 60%|██████    | 8835/14725 [4:30:32<2:56:26,  1.80s/it]
 60%|██████    | 8836/14725 [4:30:37<4:26:38,  2.72s/it]
 60%|██████    | 8837/14725 [4:30:39<3:59:47,  2.44s/it]
 60%|██████    | 8838/14725 [4:30:41<3:42:11,  2.26s/it]
 60%|██████    | 8839/14725 [4:30:42<3:29:15,  2.13s/it]
 60%|██████    | 8840/14725 [4:30:44<3:22:34,  2.07s/it]
                                                        

 60%|██████    | 8840/14725 [4:30:44<3:22:34,  2.07s/it]
 60%|██████    | 8841/14725 [4:30:46<3:15:06,  1.99s/it]
 60%|██████    | 8842/14725 [4:30:48<3:08:19,  1.92s/it]
 60%|███
+0: {'loss': 0.8101, 'grad_norm': 0.7192635384944062, 'learning_rate': 2e-05, 'epoch': 0.6}
+0: ███    | 8843/14725 [4:30:50<3:03:45,  1.87s/it]
 60%|██████    | 8844/14725 [4:30:51<3:00:53,  1.85s/it]
 60%|██████    | 8845/14725 [4:30:53<2:59:18,  1.83s/it]
 60%|██████    | 8846/14725 [4:30:55<2:57:01,  1.81s/it]
 60%|██████    | 8847/14725 [4:30:57<2:55:43,  1.79s/it]
 60%|██████    | 8848/14725 [4:30:59<2:54:58,  1.79s/it]
 60%|██████    | 8849/14725 [4:31:00<2:56:05,  1.80s/it]
 60%|██████    | 8850/14725 [4:31:02<2:55:43,  1.79s/it]
                                                        

 60%|██████    | 8850/14725 [4:31:02<2:55:43,  1.79s/it]
 60%|██████    | 8851/14725 [4:31:04<2:56:23,  1.80s/it]
 60%|██████    | 8852/14725 [4:31:06<2:56:33,  1.80s/it]
 60%|██████    | 8853/14725 [4:31:08<2:56:12,  1.80s/it]
 60%|██████    | 8854/14725 [4:31:09<2:56:01,  1.80s/it]
 60%|██████    | 8855/14725 [4:31:11<2:54:58,  1.79s/it]
 60%|███
+0: {'loss': 0.7969, 'grad_norm': 0.8134199523772859, 'learning_rate': 2e-05, 'epoch': 0.6}
+0: ███    | 8856/14725 [4:31:13<2:58:26,  1.82s/it]
 60%|██████    | 8857/14725 [4:31:15<2:59:10,  1.83s/it]
 60%|██████    | 8858/14725 [4:31:17<2:58:29,  1.83s/it]
 60%|██████    | 8859/14725 [4:31:19<3:00:16,  1.84s/it]
 60%|██████    | 8860/14725 [4:31:20<2:59:55,  1.84s/it]
                                                        

 60%|██████    | 8860/14725 [4:31:20<2:59:55,  1.84s/it]
 60%|██████    | 8861/14725 [4:31:22<2:59:52,  1.84s/it]
 60%|██████    | 8862/14725 [4:31:24<2:57:11,  1.81s/it]
 60%|██████    | 8863/14725 [4:31:26<2:57:26,  1.82s/it]
 60%|██████    | 8864/14725 [4:31:28<2:57:51,  1.82s/it]
 60%|██████    | 8865/14725 [4:31:29<2:56:05,  1.80s/it]
 60%|██████    | 8866/14725 [4:31:31<2:56:01,  1.80s/it]
 60%|██████    | 8867/14725 [4:31:33<2:56:02,  1.80s/it]
 60%|██████    | 8868/14725 [4:31:35<2:57:12,  1.82s/it]
 60%|███
+0: {'loss': 0.808, 'grad_norm': 0.653977262064959, 'learning_rate': 2e-05, 'epoch': 0.6}
+0: {'loss': 0.8026, 'grad_norm': 0.6827831204272259, 'learning_rate': 2e-05, 'epoch': 0.6}
+0: ███    | 8869/14725 [4:31:37<2:56:04,  1.80s/it]
 60%|██████    | 8870/14725 [4:31:38<2:55:04,  1.79s/it]
                                                        

 60%|██████    | 8870/14725 [4:31:38<2:55:04,  1.79s/it]
 60%|██████    | 8871/14725 [4:31:40<2:55:16,  1.80s/it]
 60%|██████    | 8872/14725 [4:31:42<2:55:35,  1.80s/it]
 60%|██████    | 8873/14725 [4:31:44<2:58:05,  1.83s/it]
 60%|██████    | 8874/14725 [4:31:46<2:56:06,  1.81s/it]
 60%|██████    | 8875/14725 [4:31:47<2:55:49,  1.80s/it]
 60%|██████    | 8876/14725 [4:31:49<2:57:51,  1.82s/it]
 60%|██████    | 8877/14725 [4:31:51<2:55:59,  1.81s/it]
 60%|██████    | 8878/14725 [4:31:53<2:54:39,  1.79s/it]
 60%|██████    | 8879/14725 [4:31:55<2:57:57,  1.83s/it]
 60%|██████    | 8880/14725 [4:31:57<2:58:16,  1.83s/it]
                                                        

 60%|██████  
+0: {'loss': 0.801, 'grad_norm': 0.7007968182880768, 'learning_rate': 2e-05, 'epoch': 0.6}
+0:   | 8880/14725 [4:31:57<2:58:16,  1.83s/it]
 60%|██████    | 8881/14725 [4:31:58<2:57:38,  1.82s/it]
 60%|██████    | 8882/14725 [4:32:00<2:55:56,  1.81s/it]
 60%|██████    | 8883/14725 [4:32:02<2:54:36,  1.79s/it]
 60%|██████    | 8884/14725 [4:32:04<2:54:13,  1.79s/it]
 60%|██████    | 8885/14725 [4:32:06<2:58:35,  1.83s/it]
 60%|██████    | 8886/14725 [4:32:07<2:56:40,  1.82s/it]
 60%|██████    | 8887/14725 [4:32:09<2:56:54,  1.82s/it]
 60%|██████    | 8888/14725 [4:32:11<2:57:36,  1.83s/it]
 60%|██████    | 8889/14725 [4:32:13<2:55:28,  1.80s/it]
 60%|██████    | 8890/14725 [4:32:15<2:56:44,  1.82s/it]
                                                        

 60%|██████    | 8890/14725 [4:32:15<2:56:44,  1.82s/it]
 60%|██████    | 8891/14725 [4:32:17<2:57:06,  1.82s/it]
 60%|██████    | 8892/14725 [4:32:19<3:02:58,  1.88s/it]
 60%|██████  
+0: {'loss': 0.8458, 'grad_norm': 0.801112125917302, 'learning_rate': 2e-05, 'epoch': 0.6}
+0:   | 8893/14725 [4:32:20<3:01:42,  1.87s/it]
 60%|██████    | 8894/14725 [4:32:22<3:01:20,  1.87s/it]
 60%|██████    | 8895/14725 [4:32:24<2:58:41,  1.84s/it]
 60%|██████    | 8896/14725 [4:32:26<2:57:15,  1.82s/it]
 60%|██████    | 8897/14725 [4:32:28<2:56:56,  1.82s/it]
 60%|██████    | 8898/14725 [4:32:29<2:56:45,  1.82s/it]
 60%|██████    | 8899/14725 [4:32:31<2:57:11,  1.82s/it]
 60%|██████    | 8900/14725 [4:32:33<2:58:38,  1.84s/it]
                                                        

 60%|██████    | 8900/14725 [4:32:33<2:58:38,  1.84s/it]
 60%|██████    | 8901/14725 [4:32:35<3:00:51,  1.86s/it]
 60%|██████    | 8902/14725 [4:32:37<2:58:04,  1.83s/it]
 60%|██████    | 8903/14725 [4:32:39<2:57:10,  1.83s/it]
 60%|██████    | 8904/14725 [4:32:40<2:57:41,  1.83s/it]
 60%|██████    | 8905/14725 [4:32:42<2:56:58,  1.82s/it]
 60%|██████  
+0: {'loss': 0.8058, 'grad_norm': 0.7349031436999375, 'learning_rate': 2e-05, 'epoch': 0.61}
+0:   | 8906/14725 [4:32:44<2:55:07,  1.81s/it]
 60%|██████    | 8907/14725 [4:32:46<2:55:06,  1.81s/it]
 60%|██████    | 8908/14725 [4:32:48<2:56:49,  1.82s/it]
 61%|██████    | 8909/14725 [4:32:49<2:55:16,  1.81s/it]
 61%|██████    | 8910/14725 [4:32:51<2:55:58,  1.82s/it]
                                                        

 61%|██████    | 8910/14725 [4:32:51<2:55:58,  1.82s/it]
 61%|██████    | 8911/14725 [4:32:53<2:55:11,  1.81s/it]
 61%|██████    | 8912/14725 [4:32:55<2:53:45,  1.79s/it]
 61%|██████    | 8913/14725 [4:32:57<2:54:21,  1.80s/it]
 61%|██████    | 8914/14725 [4:32:58<2:54:12,  1.80s/it]
 61%|██████    | 8915/14725 [4:33:00<2:55:44,  1.81s/it]
 61%|██████    | 8916/14725 [4:33:02<2:55:05,  1.81s/it]
 61%|██████    | 8917/14725 [4:33:04<2:58:17,  1.84s/it]
 61%|██████    | 8918/14725 [4:33:06<2:56:23,  1.82s/it]
 61%|██████  
+0: {'loss': 0.8096, 'grad_norm': 0.6617844303801361, 'learning_rate': 2e-05, 'epoch': 0.61}
+0: {'loss': 0.8054, 'grad_norm': 0.76281404190869, 'learning_rate': 2e-05, 'epoch': 0.61}
+0:   | 8919/14725 [4:33:08<2:56:13,  1.82s/it]
 61%|██████    | 8920/14725 [4:33:09<2:55:25,  1.81s/it]
                                                        

 61%|██████    | 8920/14725 [4:33:09<2:55:25,  1.81s/it]
 61%|██████    | 8921/14725 [4:33:11<2:56:25,  1.82s/it]
 61%|██████    | 8922/14725 [4:33:13<2:56:11,  1.82s/it]
 61%|██████    | 8923/14725 [4:33:15<2:54:42,  1.81s/it]
 61%|██████    | 8924/14725 [4:33:17<2:54:43,  1.81s/it]
 61%|██████    | 8925/14725 [4:33:18<2:53:37,  1.80s/it]
 61%|██████    | 8926/14725 [4:33:20<2:52:47,  1.79s/it]
 61%|██████    | 8927/14725 [4:33:22<2:55:11,  1.81s/it]
 61%|██████    | 8928/14725 [4:33:24<2:53:10,  1.79s/it]
 61%|██████    | 8929/14725 [4:33:26<2:57:25,  1.84s/it]
 61%|██████    | 8930/14725 [4:33:28<3:00:06,  1.86s/it]
                                                        

 61%|██████    | 8930/14
+0: {'loss': 0.8133, 'grad_norm': 0.6892078492241446, 'learning_rate': 2e-05, 'epoch': 0.61}
+0: 725 [4:33:28<3:00:06,  1.86s/it]
 61%|██████    | 8931/14725 [4:33:30<2:58:55,  1.85s/it]
 61%|██████    | 8932/14725 [4:33:31<2:58:47,  1.85s/it]
 61%|██████    | 8933/14725 [4:33:33<2:55:43,  1.82s/it]
 61%|██████    | 8934/14725 [4:33:35<2:53:39,  1.80s/it]
 61%|██████    | 8935/14725 [4:33:37<2:53:28,  1.80s/it]
 61%|██████    | 8936/14725 [4:33:38<2:52:43,  1.79s/it]
 61%|██████    | 8937/14725 [4:33:40<2:54:50,  1.81s/it]
 61%|██████    | 8938/14725 [4:33:42<2:55:58,  1.82s/it]
 61%|██████    | 8939/14725 [4:33:44<2:55:09,  1.82s/it]
 61%|██████    | 8940/14725 [4:33:46<2:54:09,  1.81s/it]
                                                        

 61%|██████    | 8940/14725 [4:33:46<2:54:09,  1.81s/it]
 61%|██████    | 8941/14725 [4:33:48<2:53:11,  1.80s/it]
 61%|██████    | 8942/14725 [4:33:49<2:52:13,  1.79s/it]
 61%|██████    | 8943/14
+0: {'loss': 0.8161, 'grad_norm': 0.6983820980568354, 'learning_rate': 2e-05, 'epoch': 0.61}
+0: 725 [4:33:51<2:52:55,  1.79s/it]
 61%|██████    | 8944/14725 [4:33:53<2:53:24,  1.80s/it]
 61%|██████    | 8945/14725 [4:33:55<2:55:36,  1.82s/it]
 61%|██████    | 8946/14725 [4:33:57<2:54:20,  1.81s/it]
 61%|██████    | 8947/14725 [4:33:58<2:52:44,  1.79s/it]
 61%|██████    | 8948/14725 [4:34:00<2:52:53,  1.80s/it]
 61%|██████    | 8949/14725 [4:34:02<2:52:34,  1.79s/it]
 61%|██████    | 8950/14725 [4:34:04<2:51:38,  1.78s/it]
                                                        

 61%|██████    | 8950/14725 [4:34:04<2:51:38,  1.78s/it]
 61%|██████    | 8951/14725 [4:34:06<2:54:09,  1.81s/it]
 61%|██████    | 8952/14725 [4:34:07<2:53:54,  1.81s/it]
 61%|██████    | 8953/14725 [4:34:09<2:54:12,  1.81s/it]
 61%|██████    | 8954/14725 [4:34:11<2:54:20,  1.81s/it]
 61%|██████    | 8955/14725 [4:34:13<2:54:20,  1.81s/it]
 61%|██████    | 8956/14
+0: {'loss': 0.8073, 'grad_norm': 0.6954444031395811, 'learning_rate': 2e-05, 'epoch': 0.61}
+0: 725 [4:34:15<2:55:06,  1.82s/it]
 61%|██████    | 8957/14725 [4:34:16<2:54:34,  1.82s/it]
 61%|██████    | 8958/14725 [4:34:18<2:53:10,  1.80s/it]
 61%|██████    | 8959/14725 [4:34:20<2:52:20,  1.79s/it]
 61%|██████    | 8960/14725 [4:34:22<2:53:00,  1.80s/it]
                                                        

 61%|██████    | 8960/14725 [4:34:22<2:53:00,  1.80s/it]
 61%|██████    | 8961/14725 [4:34:24<2:51:55,  1.79s/it]
 61%|██████    | 8962/14725 [4:34:25<2:51:54,  1.79s/it]
 61%|██████    | 8963/14725 [4:34:27<2:52:24,  1.80s/it]
 61%|██████    | 8964/14725 [4:34:29<2:51:26,  1.79s/it]
 61%|██████    | 8965/14725 [4:34:31<2:51:05,  1.78s/it]
 61%|██████    | 8966/14725 [4:34:33<2:53:51,  1.81s/it]
 61%|██████    | 8967/14725 [4:34:34<2:53:09,  1.80s/it]
 61%|██████    | 8968/14725 [4:34:36<2:51:16,  1.79s/it]
 61%|██████    | 8969/14
+0: {'loss': 0.8055, 'grad_norm': 0.7157511433844278, 'learning_rate': 2e-05, 'epoch': 0.61}
+0: {'loss': 0.8156, 'grad_norm': 0.7060914169311566, 'learning_rate': 2e-05, 'epoch': 0.61}
+0: 725 [4:34:38<2:50:25,  1.78s/it]
 61%|██████    | 8970/14725 [4:34:40<2:52:15,  1.80s/it]
                                                        

 61%|██████    | 8970/14725 [4:34:40<2:52:15,  1.80s/it]
 61%|██████    | 8971/14725 [4:34:42<2:52:27,  1.80s/it]
 61%|██████    | 8972/14725 [4:34:43<2:51:01,  1.78s/it]
 61%|██████    | 8973/14725 [4:34:45<2:50:00,  1.77s/it]
 61%|██████    | 8974/14725 [4:34:47<2:49:00,  1.76s/it]
 61%|██████    | 8975/14725 [4:34:49<2:49:03,  1.76s/it]
 61%|██████    | 8976/14725 [4:34:50<2:51:18,  1.79s/it]
 61%|██████    | 8977/14725 [4:34:52<2:50:25,  1.78s/it]
 61%|██████    | 8978/14725 [4:34:54<2:49:50,  1.77s/it]
 61%|██████    | 8979/14725 [4:34:56<2:50:06,  1.78s/it]
 61%|██████    | 8980/14725 [4:34:57<2:51:33,  1.79s/it]
                                                        

 61%|██████    | 8980/14725 [4:34:5
+0: {'loss': 0.8234, 'grad_norm': 0.660643379561488, 'learning_rate': 2e-05, 'epoch': 0.61}
+0: 7<2:51:33,  1.79s/it]
 61%|██████    | 8981/14725 [4:34:59<2:51:42,  1.79s/it]
 61%|██████    | 8982/14725 [4:35:01<2:50:51,  1.79s/it]
 61%|██████    | 8983/14725 [4:35:03<2:51:20,  1.79s/it]
 61%|██████    | 8984/14725 [4:35:05<2:55:39,  1.84s/it]
 61%|██████    | 8985/14725 [4:35:07<2:54:35,  1.83s/it]
 61%|██████    | 8986/14725 [4:35:08<2:52:22,  1.80s/it]
 61%|██████    | 8987/14725 [4:35:10<2:57:10,  1.85s/it]
 61%|██████    | 8988/14725 [4:35:12<2:55:23,  1.83s/it]
 61%|██████    | 8989/14725 [4:35:14<2:53:20,  1.81s/it]
 61%|██████    | 8990/14725 [4:35:16<2:53:33,  1.82s/it]
                                                        

 61%|██████    | 8990/14725 [4:35:16<2:53:33,  1.82s/it]
 61%|██████    | 8991/14725 [4:35:18<2:55:19,  1.83s/it]
 61%|██████    | 8992/14725 [4:35:19<2:53:22,  1.81s/it]
 61%|██████    | 8993/14725 [4:35:2
+0: {'loss': 0.8361, 'grad_norm': 0.7621295982474321, 'learning_rate': 2e-05, 'epoch': 0.61}
+0: 1<2:53:17,  1.81s/it]
 61%|██████    | 8994/14725 [4:35:23<2:53:11,  1.81s/it]
 61%|██████    | 8995/14725 [4:35:25<2:53:41,  1.82s/it]
 61%|██████    | 8996/14725 [4:35:27<2:54:32,  1.83s/it]
 61%|██████    | 8997/14725 [4:35:28<2:54:50,  1.83s/it]
 61%|██████    | 8998/14725 [4:35:30<2:54:19,  1.83s/it]
 61%|██████    | 8999/14725 [4:35:32<2:53:40,  1.82s/it]
 61%|██████    | 9000/14725 [4:35:34<2:51:25,  1.80s/it]
                                                        

 61%|██████    | 9000/14725 [4:35:34<2:51:25,  1.80s/it]
 61%|██████    | 9001/14725 [4:35:36<2:54:34,  1.83s/it]
 61%|██████    | 9002/14725 [4:35:38<2:55:15,  1.84s/it]
 61%|██████    | 9003/14725 [4:35:39<2:55:44,  1.84s/it]
 61%|██████    | 9004/14725 [4:35:41<2:53:56,  1.82s/it]
 61%|██████    | 9005/14725 [4:35:43<2:53:19,  1.82s/it]
 61%|██████    | 9006/14725 [4:35:4
+0: {'loss': 0.8193, 'grad_norm': 0.7022813483819133, 'learning_rate': 2e-05, 'epoch': 0.61}
+0: 5<2:53:55,  1.82s/it]
 61%|██████    | 9007/14725 [4:35:47<2:54:39,  1.83s/it]
 61%|██████    | 9008/14725 [4:35:48<2:52:10,  1.81s/it]
 61%|██████    | 9009/14725 [4:35:50<2:53:42,  1.82s/it]
 61%|██████    | 9010/14725 [4:35:52<2:56:50,  1.86s/it]
                                                        

 61%|██████    | 9010/14725 [4:35:52<2:56:50,  1.86s/it]
 61%|██████    | 9011/14725 [4:35:54<2:54:05,  1.83s/it]
 61%|██████    | 9012/14725 [4:35:56<2:51:59,  1.81s/it]
 61%|██████    | 9013/14725 [4:35:58<2:50:38,  1.79s/it]
 61%|██████    | 9014/14725 [4:35:59<2:51:32,  1.80s/it]
 61%|██████    | 9015/14725 [4:36:01<2:49:44,  1.78s/it]
 61%|██████    | 9016/14725 [4:36:03<2:50:07,  1.79s/it]
 61%|██████    | 9017/14725 [4:36:05<2:49:43,  1.78s/it]
 61%|██████    | 9018/14725 [4:36:06<2:49:13,  1.78s/it]
 61%|██████    | 9019/14725 [4:36:0
+0: {'loss': 0.7942, 'grad_norm': 0.7115737285919298, 'learning_rate': 2e-05, 'epoch': 0.61}
+0: {'loss': 0.7893, 'grad_norm': 0.725995403366021, 'learning_rate': 2e-05, 'epoch': 0.61}
+0: 8<2:49:00,  1.78s/it]
 61%|██████▏   | 9020/14725 [4:36:10<2:51:15,  1.80s/it]
                                                        

 61%|██████▏   | 9020/14725 [4:36:10<2:51:15,  1.80s/it]
 61%|██████▏   | 9021/14725 [4:36:12<2:51:42,  1.81s/it]
 61%|██████▏   | 9022/14725 [4:36:14<2:49:55,  1.79s/it]
 61%|██████▏   | 9023/14725 [4:36:15<2:50:12,  1.79s/it]
 61%|██████▏   | 9024/14725 [4:36:17<2:50:13,  1.79s/it]
 61%|██████▏   | 9025/14725 [4:36:19<2:50:33,  1.80s/it]
 61%|██████▏   | 9026/14725 [4:36:21<2:50:08,  1.79s/it]
 61%|██████▏   | 9027/14725 [4:36:23<2:48:33,  1.77s/it]
 61%|██████▏   | 9028/14725 [4:36:24<2:50:36,  1.80s/it]
 61%|██████▏   | 9029/14725 [4:36:26<2:49:33,  1.79s/it]
 61%|██████▏   | 9030/14725 [4:36:28<2:51:48,  1.81s/it]
                                                        

 61%|██████▏   | 903
+0: {'loss': 0.8016, 'grad_norm': 0.7252387615832564, 'learning_rate': 2e-05, 'epoch': 0.61}
+0: 0/14725 [4:36:28<2:51:48,  1.81s/it]
 61%|██████▏   | 9031/14725 [4:36:30<2:53:16,  1.83s/it]
 61%|██████▏   | 9032/14725 [4:36:32<2:52:41,  1.82s/it]
 61%|██████▏   | 9033/14725 [4:36:34<2:53:42,  1.83s/it]
 61%|██████▏   | 9034/14725 [4:36:35<2:54:31,  1.84s/it]
 61%|██████▏   | 9035/14725 [4:36:37<2:52:34,  1.82s/it]
 61%|██████▏   | 9036/14725 [4:36:39<2:52:12,  1.82s/it]
 61%|██████▏   | 9037/14725 [4:36:41<2:52:04,  1.82s/it]
 61%|██████▏   | 9038/14725 [4:36:43<2:53:29,  1.83s/it]
 61%|██████▏   | 9039/14725 [4:36:45<2:53:14,  1.83s/it]
 61%|██████▏   | 9040/14725 [4:36:46<2:53:02,  1.83s/it]
                                                        

 61%|██████▏   | 9040/14725 [4:36:46<2:53:02,  1.83s/it]
 61%|██████▏   | 9041/14725 [4:36:48<2:54:34,  1.84s/it]
 61%|██████▏   | 9042/14725 [4:36:50<2:52:05,  1.82s/it]
 61%|�
+0: {'loss': 0.8309, 'grad_norm': 0.6879551621415131, 'learning_rate': 2e-05, 'epoch': 0.61}
+0: ��█████▏   | 9043/14725 [4:36:52<2:52:45,  1.82s/it]
 61%|██████▏   | 9044/14725 [4:36:54<2:54:01,  1.84s/it]
 61%|██████▏   | 9045/14725 [4:36:55<2:52:55,  1.83s/it]
 61%|██████▏   | 9046/14725 [4:36:57<2:53:23,  1.83s/it]
 61%|██████▏   | 9047/14725 [4:36:59<2:52:55,  1.83s/it]
 61%|██████▏   | 9048/14725 [4:37:01<2:54:04,  1.84s/it]
 61%|██████▏   | 9049/14725 [4:37:03<2:51:44,  1.82s/it]
 61%|██████▏   | 9050/14725 [4:37:05<2:52:23,  1.82s/it]
                                                        

 61%|██████▏   | 9050/14725 [4:37:05<2:52:23,  1.82s/it]
 61%|██████▏   | 9051/14725 [4:37:06<2:51:01,  1.81s/it]
 61%|██████▏   | 9052/14725 [4:37:08<2:50:02,  1.80s/it]
 61%|██████▏   | 9053/14725 [4:37:10<2:50:09,  1.80s/it]
 61%|██████▏   | 9054/14725 [4:37:12<2:49:09,  1.79s/it]
 61%|██████▏   | 9055/14725 [4:37:1
+0: {'loss': 0.7988, 'grad_norm': 0.7231285059022727, 'learning_rate': 2e-05, 'epoch': 0.62}
+0: 4<2:49:08,  1.79s/it]
 62%|██████▏   | 9056/14725 [4:37:15<2:48:47,  1.79s/it]
 62%|██████▏   | 9057/14725 [4:37:17<2:48:34,  1.78s/it]
 62%|██████▏   | 9058/14725 [4:37:19<2:47:48,  1.78s/it]
 62%|██████▏   | 9059/14725 [4:37:21<2:47:45,  1.78s/it]
 62%|██████▏   | 9060/14725 [4:37:22<2:48:39,  1.79s/it]
                                                        

 62%|██████▏   | 9060/14725 [4:37:22<2:48:39,  1.79s/it]
 62%|██████▏   | 9061/14725 [4:37:24<2:48:21,  1.78s/it]
 62%|██████▏   | 9062/14725 [4:37:26<2:47:44,  1.78s/it]
 62%|██████▏   | 9063/14725 [4:37:28<2:48:09,  1.78s/it]
 62%|██████▏   | 9064/14725 [4:37:30<2:47:44,  1.78s/it]
 62%|██████▏   | 9065/14725 [4:37:31<2:47:04,  1.77s/it]
 62%|██████▏   | 9066/14725 [4:37:33<2:47:36,  1.78s/it]
 62%|██████▏   | 9067/14725 [4:37:35<2:48:05,  1.78s/it]
 62%|█████�
+0: {'loss': 0.8025, 'grad_norm': 0.6647578039081774, 'learning_rate': 2e-05, 'epoch': 0.62}
+0: ��▏   | 9068/14725 [4:37:37<2:47:55,  1.78s/it]
 62%|██████▏   | 9069/14725 [4:37:38<2:49:06,  1.79s/it]
 62%|██████▏   | 9070/14725 [4:37:40<2:48:29,  1.79s/it]
                                                        

 62%|██████▏   | 9070/14725 [4:37:40<2:48:29,  1.79s/it]
 62%|██████▏   | 9071/14725 [4:37:42<2:49:17,  1.80s/it]
 62%|██████▏   | 9072/14725 [4:37:44<2:49:24,  1.80s/it]
 62%|██████▏   | 9073/14725 [4:37:46<2:50:08,  1.81s/it]
 62%|██████▏   | 9074/14725 [4:37:47<2:49:27,  1.80s/it]
 62%|██████▏   | 9075/14725 [4:37:49<2:48:59,  1.79s/it]
 62%|██████▏   | 9076/14725 [4:37:51<2:47:45,  1.78s/it]
 62%|██████▏   | 9077/14725 [4:37:53<2:48:30,  1.79s/it]
 62%|██████▏   | 9078/14725 [4:37:55<2:49:50,  1.80s/it]
 62%|██████▏   | 9079/14725 [4:37:56<2:47:53,  1.78s/it]
 62%|██████▏   | 9080/14725 [4:37:58<2:48:00,  1.7
+0: {'loss': 0.7954, 'grad_norm': 0.7408147101732155, 'learning_rate': 2e-05, 'epoch': 0.62}
+0: {'loss': 0.8116, 'grad_norm': 0.7265692197468988, 'learning_rate': 2e-05, 'epoch': 0.62}
+0: 9s/it]
                                                        

 62%|██████▏   | 9080/14725 [4:37:58<2:48:00,  1.79s/it]
 62%|██████▏   | 9081/14725 [4:38:00<2:47:33,  1.78s/it]
 62%|██████▏   | 9082/14725 [4:38:02<2:49:56,  1.81s/it]
 62%|██████▏   | 9083/14725 [4:38:04<2:51:07,  1.82s/it]
 62%|██████▏   | 9084/14725 [4:38:05<2:50:32,  1.81s/it]
 62%|██████▏   | 9085/14725 [4:38:07<2:49:44,  1.81s/it]
 62%|██████▏   | 9086/14725 [4:38:09<2:49:25,  1.80s/it]
 62%|██████▏   | 9087/14725 [4:38:11<2:48:17,  1.79s/it]
 62%|██████▏   | 9088/14725 [4:38:13<2:48:14,  1.79s/it]
 62%|██████▏   | 9089/14725 [4:38:14<2:48:29,  1.79s/it]
 62%|██████▏   | 9090/14725 [4:38:16<2:47:44,  1.79s/it]
                                                        

 62%|██████▏   | 9090/14725 [4:38:16<2:47:44,  1.79s/it]
 62%|██████▏   | 9091/14725 [4:38:1
+0: {'loss': 0.8088, 'grad_norm': 0.7052421858244635, 'learning_rate': 2e-05, 'epoch': 0.62}
+0: 8<2:48:45,  1.80s/it]
 62%|██████▏   | 9092/14725 [4:38:20<2:49:26,  1.80s/it]
 62%|██████▏   | 9093/14725 [4:38:22<2:49:29,  1.81s/it]
 62%|██████▏   | 9094/14725 [4:38:23<2:49:46,  1.81s/it]
 62%|██████▏   | 9095/14725 [4:38:25<2:48:50,  1.80s/it]
 62%|██████▏   | 9096/14725 [4:38:27<2:48:39,  1.80s/it]
 62%|██████▏   | 9097/14725 [4:38:29<2:48:59,  1.80s/it]
 62%|██████▏   | 9098/14725 [4:38:31<2:48:26,  1.80s/it]
 62%|██████▏   | 9099/14725 [4:38:32<2:48:32,  1.80s/it]
 62%|██████▏   | 9100/14725 [4:38:34<2:48:14,  1.79s/it]
                                                        

 62%|██████▏   | 9100/14725 [4:38:34<2:48:14,  1.79s/it]
 62%|██████▏   | 9101/14725 [4:38:36<2:48:04,  1.79s/it]
 62%|██████▏   | 9102/14725 [4:38:38<2:48:38,  1.80s/it]
 62%|██████▏   | 9103/14725 [4:38:40<2:50:45,  1.82s/it]
 62%|█████�
+0: {'loss': 0.8172, 'grad_norm': 0.7379396896811866, 'learning_rate': 2e-05, 'epoch': 0.62}
+0: ��▏   | 9104/14725 [4:38:41<2:48:59,  1.80s/it]
 62%|██████▏   | 9105/14725 [4:38:43<2:50:11,  1.82s/it]
 62%|██████▏   | 9106/14725 [4:38:45<2:48:29,  1.80s/it]
 62%|██████▏   | 9107/14725 [4:38:47<2:47:49,  1.79s/it]
 62%|██████▏   | 9108/14725 [4:38:49<2:48:09,  1.80s/it]
 62%|██████▏   | 9109/14725 [4:38:50<2:47:57,  1.79s/it]
 62%|██████▏   | 9110/14725 [4:38:52<2:46:47,  1.78s/it]
                                                        

 62%|██████▏   | 9110/14725 [4:38:52<2:46:47,  1.78s/it]
 62%|██████▏   | 9111/14725 [4:38:54<2:48:08,  1.80s/it]
 62%|██████▏   | 9112/14725 [4:38:56<2:48:01,  1.80s/it]
 62%|██████▏   | 9113/14725 [4:38:58<2:49:49,  1.82s/it]
 62%|██████▏   | 9114/14725 [4:38:59<2:50:27,  1.82s/it]
 62%|██████▏   | 9115/14725 [4:39:01<2:49:26,  1.81s/it]
 62%|██████▏   | 9116/14725 [4:39:03<2:49:19,  1.8
+0: {'loss': 0.7993, 'grad_norm': 0.6919588803061918, 'learning_rate': 2e-05, 'epoch': 0.62}
+0: 1s/it]
 62%|██████▏   | 9117/14725 [4:39:05<2:51:55,  1.84s/it]
 62%|██████▏   | 9118/14725 [4:39:07<2:49:42,  1.82s/it]
 62%|██████▏   | 9119/14725 [4:39:09<2:49:59,  1.82s/it]
 62%|██████▏   | 9120/14725 [4:39:10<2:48:51,  1.81s/it]
                                                        

 62%|██████▏   | 9120/14725 [4:39:10<2:48:51,  1.81s/it]
 62%|██████▏   | 9121/14725 [4:39:12<2:49:41,  1.82s/it]
 62%|██████▏   | 9122/14725 [4:39:14<2:50:30,  1.83s/it]
 62%|██████▏   | 9123/14725 [4:39:16<2:50:40,  1.83s/it]
 62%|██████▏   | 9124/14725 [4:39:18<2:53:22,  1.86s/it]
 62%|██████▏   | 9125/14725 [4:39:20<2:51:08,  1.83s/it]
 62%|██████▏   | 9126/14725 [4:39:21<2:48:20,  1.80s/it]
 62%|██████▏   | 9127/14725 [4:39:23<2:48:32,  1.81s/it]
 62%|██████▏   | 9128/14725 [4:39:25<2:48:08,  1.80s/it]
 62%|██████▏   | 9129/
+0: {'loss': 0.8251, 'grad_norm': 0.7219216092882166, 'learning_rate': 2e-05, 'epoch': 0.62}
+0: {'loss': 0.8366, 'grad_norm': 0.7058569914156279, 'learning_rate': 2e-05, 'epoch': 0.62}
+0: 14725 [4:39:27<2:49:45,  1.82s/it]
 62%|██████▏   | 9130/14725 [4:39:29<2:47:56,  1.80s/it]
                                                        

 62%|██████▏   | 9130/14725 [4:39:29<2:47:56,  1.80s/it]
 62%|██████▏   | 9131/14725 [4:39:30<2:50:04,  1.82s/it]
 62%|██████▏   | 9132/14725 [4:39:32<2:49:24,  1.82s/it]
 62%|██████▏   | 9133/14725 [4:39:34<2:48:16,  1.81s/it]
 62%|██████▏   | 9134/14725 [4:39:36<2:47:24,  1.80s/it]
 62%|██████▏   | 9135/14725 [4:39:38<2:47:21,  1.80s/it]
 62%|██████▏   | 9136/14725 [4:39:39<2:46:48,  1.79s/it]
 62%|██████▏   | 9137/14725 [4:39:41<2:48:19,  1.81s/it]
 62%|██████▏   | 9138/14725 [4:39:43<2:48:42,  1.81s/it]
 62%|██████▏   | 9139/14725 [4:39:45<2:46:49,  1.79s/it]
 62%|██████▏   | 9140/14725 [4:39:47<2:52:44,  1.86s/it]
                                                        

 62%|█████�
+0: {'loss': 0.8108, 'grad_norm': 0.6652042877015149, 'learning_rate': 2e-05, 'epoch': 0.62}
+0: ��▏   | 9140/14725 [4:39:47<2:52:44,  1.86s/it]
 62%|██████▏   | 9141/14725 [4:39:49<2:53:18,  1.86s/it]
 62%|██████▏   | 9142/14725 [4:39:51<2:54:24,  1.87s/it]
 62%|██████▏   | 9143/14725 [4:39:52<2:52:31,  1.85s/it]
 62%|██████▏   | 9144/14725 [4:39:54<2:55:40,  1.89s/it]
 62%|██████▏   | 9145/14725 [4:39:56<2:53:31,  1.87s/it]
 62%|██████▏   | 9146/14725 [4:39:58<2:52:58,  1.86s/it]
 62%|██████▏   | 9147/14725 [4:40:00<2:53:23,  1.87s/it]
 62%|██████▏   | 9148/14725 [4:40:02<2:50:34,  1.84s/it]
 62%|██████▏   | 9149/14725 [4:40:03<2:48:34,  1.81s/it]
 62%|██████▏   | 9150/14725 [4:40:05<2:47:11,  1.80s/it]
                                                        

 62%|██████▏   | 9150/14725 [4:40:05<2:47:11,  1.80s/it]
 62%|██████▏   | 9151/14725 [4:40:07<2:48:00,  1.81s/it]
 62%|██████▏   | 9152/14725 [4:40:09<2:46:49,  1.8
+0: {'loss': 0.8064, 'grad_norm': 0.6313428408393614, 'learning_rate': 2e-05, 'epoch': 0.62}
+0: 0s/it]
 62%|██████▏   | 9153/14725 [4:40:11<2:46:10,  1.79s/it]
 62%|██████▏   | 9154/14725 [4:40:12<2:44:46,  1.77s/it]
 62%|██████▏   | 9155/14725 [4:40:14<2:46:11,  1.79s/it]
 62%|██████▏   | 9156/14725 [4:40:16<2:44:51,  1.78s/it]
 62%|██████▏   | 9157/14725 [4:40:18<2:44:09,  1.77s/it]
 62%|██████▏   | 9158/14725 [4:40:19<2:47:41,  1.81s/it]
 62%|██████▏   | 9159/14725 [4:40:21<2:46:47,  1.80s/it]
 62%|██████▏   | 9160/14725 [4:40:23<2:52:41,  1.86s/it]
                                                        

 62%|██████▏   | 9160/14725 [4:40:23<2:52:41,  1.86s/it]
 62%|██████▏   | 9161/14725 [4:40:25<2:53:26,  1.87s/it]
 62%|██████▏   | 9162/14725 [4:40:27<2:52:52,  1.86s/it]
 62%|██████▏   | 9163/14725 [4:40:29<2:52:40,  1.86s/it]
 62%|██████▏   | 9164/14725 [4:40:31<2:52:23,  1.86s/it]
 62%|██████▏   | 9165/
+0: {'loss': 0.8165, 'grad_norm': 0.6703617499621259, 'learning_rate': 2e-05, 'epoch': 0.62}
+0: 14725 [4:40:33<2:51:10,  1.85s/it]
 62%|██████▏   | 9166/14725 [4:40:34<2:52:46,  1.86s/it]
 62%|██████▏   | 9167/14725 [4:40:36<2:49:59,  1.84s/it]
 62%|██████▏   | 9168/14725 [4:40:38<2:48:13,  1.82s/it]
 62%|██████▏   | 9169/14725 [4:40:40<2:48:42,  1.82s/it]
 62%|██████▏   | 9170/14725 [4:40:42<2:47:57,  1.81s/it]
                                                        

 62%|██████▏   | 9170/14725 [4:40:42<2:47:57,  1.81s/it]
 62%|██████▏   | 9171/14725 [4:40:43<2:47:01,  1.80s/it]
 62%|██████▏   | 9172/14725 [4:40:45<2:47:48,  1.81s/it]
 62%|██████▏   | 9173/14725 [4:40:47<2:53:40,  1.88s/it]
 62%|██████▏   | 9174/14725 [4:40:49<2:50:34,  1.84s/it]
 62%|██████▏   | 9175/14725 [4:40:51<2:48:53,  1.83s/it]
 62%|██��███▏   | 9176/14725 [4:40:53<2:48:22,  1.82s/it]
 62%|██████▏   | 9177/14725 [4:40:54<2:46:10,  1.80s/it]
 62%|█
+0: {'loss': 0.8161, 'grad_norm': 0.7275305197623609, 'learning_rate': 2e-05, 'epoch': 0.62}
+0: █████▏   | 9178/14725 [4:40:56<2:45:56,  1.79s/it]
 62%|██████▏   | 9179/14725 [4:40:58<2:46:01,  1.80s/it]
 62%|██████▏   | 9180/14725 [4:41:00<2:47:29,  1.81s/it]
                                                        

 62%|██████▏   | 9180/14725 [4:41:00<2:47:29,  1.81s/it]
 62%|██████▏   | 9181/14725 [4:41:02<2:48:24,  1.82s/it]
 62%|██████▏   | 9182/14725 [4:41:03<2:46:21,  1.80s/it]
 62%|██████▏   | 9183/14725 [4:41:05<2:45:13,  1.79s/it]
 62%|██████▏   | 9184/14725 [4:41:07<2:45:54,  1.80s/it]
 62%|██████▏   | 9185/14725 [4:41:09<2:47:16,  1.81s/it]
 62%|██████▏   | 9186/14725 [4:41:11<2:47:33,  1.82s/it]
 62%|██████▏   | 9187/14725 [4:41:12<2:45:32,  1.79s/it]
 62%|██████▏   | 9188/14725 [4:41:14<2:46:11,  1.80s/it]
 62%|██████▏   | 9189/14725 [4:41:16<2:45:25,  1.79s/it]
 62%|██████▏   | 9190/14725 [4:41:18<
+0: {'loss': 0.7987, 'grad_norm': 0.6962127063213269, 'learning_rate': 2e-05, 'epoch': 0.62}
+0: {'loss': 0.8089, 'grad_norm': 0.7147204046909414, 'learning_rate': 2e-05, 'epoch': 0.62}
+0: 2:49:03,  1.83s/it]
                                                        

 62%|██████▏   | 9190/14725 [4:41:18<2:49:03,  1.83s/it]
 62%|██████▏   | 9191/14725 [4:41:20<2:50:09,  1.84s/it]
 62%|██████▏   | 9192/14725 [4:41:22<2:52:11,  1.87s/it]
 62%|██████▏   | 9193/14725 [4:41:23<2:49:45,  1.84s/it]
 62%|██████▏   | 9194/14725 [4:41:25<2:47:45,  1.82s/it]
 62%|██████▏   | 9195/14725 [4:41:27<2:48:45,  1.83s/it]
 62%|██████▏   | 9196/14725 [4:41:29<2:46:34,  1.81s/it]
 62%|██████▏   | 9197/14725 [4:41:31<2:45:40,  1.80s/it]
 62%|██████▏   | 9198/14725 [4:41:32<2:46:30,  1.81s/it]
 62%|██████▏   | 9199/14725 [4:41:34<2:46:57,  1.81s/it]
 62%|██████▏   | 9200/14725 [4:41:36<2:46:00,  1.80s/it]
                                                        

 62%|██████▏   | 9200/14725 [4:41:36<2:46:00,  1.80s/it]
 62%|██████▏   | 9201/
+0: {'loss': 0.7806, 'grad_norm': 0.6527985053451506, 'learning_rate': 2e-05, 'epoch': 0.63}
+0: 14725 [4:41:38<2:46:28,  1.81s/it]
 62%|██████▏   | 9202/14725 [4:41:40<2:46:41,  1.81s/it]
 62%|██████▏   | 9203/14725 [4:41:42<2:47:07,  1.82s/it]
 63%|██████▎   | 9204/14725 [4:41:43<2:46:32,  1.81s/it]
 63%|██████▎   | 9205/14725 [4:41:45<2:47:24,  1.82s/it]
 63%|██████▎   | 9206/14725 [4:41:47<2:46:20,  1.81s/it]
 63%|██████▎   | 9207/14725 [4:41:49<2:45:35,  1.80s/it]
 63%|██████▎   | 9208/14725 [4:41:51<2:46:50,  1.81s/it]
 63%|██████▎   | 9209/14725 [4:41:52<2:46:35,  1.81s/it]
 63%|██████▎   | 9210/14725 [4:41:54<2:45:36,  1.80s/it]
                                                        

 63%|██████▎   | 9210/14725 [4:41:54<2:45:36,  1.80s/it]
 63%|██████▎   | 9211/14725 [4:41:56<2:45:39,  1.80s/it]
 63%|██████▎   | 9212/14725 [4:41:58<2:46:29,  1.81s/it]
 63%|██████▎   | 9213/14725 [4:42:00<2:47:16,  1.82s/it]
 63%|█
+0: {'loss': 0.8318, 'grad_norm': 0.708695323181548, 'learning_rate': 2e-05, 'epoch': 0.63}
+0: █████▎   | 9214/14725 [4:42:02<2:48:10,  1.83s/it]
 63%|██████▎   | 9215/14725 [4:42:03<2:47:22,  1.82s/it]
 63%|██████▎   | 9216/14725 [4:42:05<2:46:48,  1.82s/it]
 63%|██████▎   | 9217/14725 [4:42:07<2:45:20,  1.80s/it]
 63%|██████▎   | 9218/14725 [4:42:09<2:45:56,  1.81s/it]
 63%|██████▎   | 9219/14725 [4:42:11<2:47:44,  1.83s/it]
 63%|██████▎   | 9220/14725 [4:42:12<2:45:51,  1.81s/it]
                                                        

 63%|██████▎   | 9220/14725 [4:42:12<2:45:51,  1.81s/it]
 63%|██████▎   | 9221/14725 [4:42:14<2:47:18,  1.82s/it]
 63%|██████▎   | 9222/14725 [4:42:16<2:45:29,  1.80s/it]
 63%|██████▎   | 9223/14725 [4:42:18<2:47:04,  1.82s/it]
 63%|██████▎   | 9224/14725 [4:42:20<2:48:46,  1.84s/it]
 63%|██████▎   | 9225/14725 [4:42:21<2:46:07,  1.81s/it]
 63%|██████▎   | 9226/14725 [4:42:23<
+0: {'loss': 0.8069, 'grad_norm': 0.6933911288076742, 'learning_rate': 2e-05, 'epoch': 0.63}
+0: 2:46:27,  1.82s/it]
 63%|██████▎   | 9227/14725 [4:42:25<2:46:46,  1.82s/it]
 63%|██████▎   | 9228/14725 [4:42:27<2:46:55,  1.82s/it]
 63%|██████▎   | 9229/14725 [4:42:29<2:45:57,  1.81s/it]
 63%|██████▎   | 9230/14725 [4:42:31<2:46:21,  1.82s/it]
                                                        

 63%|██████▎   | 9230/14725 [4:42:31<2:46:21,  1.82s/it]
 63%|██████▎   | 9231/14725 [4:42:32<2:45:29,  1.81s/it]
 63%|██████▎   | 9232/14725 [4:42:34<2:45:14,  1.80s/it]
 63%|██████▎   | 9233/14725 [4:42:36<2:45:13,  1.81s/it]
 63%|██████▎   | 9234/14725 [4:42:38<2:45:35,  1.81s/it]
 63%|██████▎   | 9235/14725 [4:42:40<2:46:10,  1.82s/it]
 63%|██████▎   | 9236/14725 [4:42:42<2:49:59,  1.86s/it]
 63%|██████▎   | 9237/14725 [4:42:43<2:49:47,  1.86s/it]
 63%|██████▎   | 9238/14725 [4:42:45<2:48:14,  1.84s/it]
 63%|██████
+0: {'loss': 0.8248, 'grad_norm': 0.682319732004826, 'learning_rate': 2e-05, 'epoch': 0.63}
+0: {'loss': 0.8176, 'grad_norm': 0.7510383730016664, 'learning_rate': 2e-05, 'epoch': 0.63}
+0: ▎   | 9239/14725 [4:42:47<2:48:52,  1.85s/it]
 63%|██████▎   | 9240/14725 [4:42:49<2:48:39,  1.84s/it]
                                                        

 63%|██████▎   | 9240/14725 [4:42:49<2:48:39,  1.84s/it]
 63%|██████▎   | 9241/14725 [4:42:51<2:49:28,  1.85s/it]
 63%|██████▎   | 9242/14725 [4:42:53<2:48:25,  1.84s/it]
 63%|██████▎   | 9243/14725 [4:42:54<2:46:32,  1.82s/it]
 63%|██████▎   | 9244/14725 [4:42:56<2:45:57,  1.82s/it]
 63%|██████▎   | 9245/14725 [4:42:58<2:43:55,  1.79s/it]
 63%|██████▎   | 9246/14725 [4:43:00<2:45:19,  1.81s/it]
 63%|██████▎   | 9247/14725 [4:43:02<2:43:26,  1.79s/it]
 63%|██████▎   | 9248/14725 [4:43:03<2:44:46,  1.81s/it]
 63%|██████▎   | 9249/14725 [4:43:05<2:43:48,  1.79s/it]
 63%|██████▎   | 9250/14725 [4:43:07<2:44:03,  1.80s/it]
                                                        

 63%|█
+0: {'loss': 0.8145, 'grad_norm': 0.7380428200244961, 'learning_rate': 2e-05, 'epoch': 0.63}
+0: █████▎   | 9250/14725 [4:43:07<2:44:03,  1.80s/it]
 63%|██████▎   | 9251/14725 [4:43:09<2:43:37,  1.79s/it]
 63%|██████▎   | 9252/14725 [4:43:10<2:42:58,  1.79s/it]
 63%|██████▎   | 9253/14725 [4:43:12<2:45:14,  1.81s/it]
 63%|██████▎   | 9254/14725 [4:43:14<2:46:41,  1.83s/it]
 63%|██████▎   | 9255/14725 [4:43:16<2:44:51,  1.81s/it]
 63%|██████▎   | 9256/14725 [4:43:18<2:45:02,  1.81s/it]
 63%|██████▎   | 9257/14725 [4:43:20<2:44:33,  1.81s/it]
 63%|██████▎   | 9258/14725 [4:43:21<2:44:28,  1.81s/it]
 63%|██████▎   | 9259/14725 [4:43:23<2:44:18,  1.80s/it]
 63%|██████▎   | 9260/14725 [4:43:25<2:45:12,  1.81s/it]
                                                        

 63%|██████▎   | 9260/14725 [4:43:25<2:45:12,  1.81s/it]
 63%|██████▎   | 9261/14725 [4:43:27<2:45:02,  1.81s/it]
 63%|██████▎   | 9262/14725 [4:43:29<
+0: {'loss': 0.8073, 'grad_norm': 0.6691411100697662, 'learning_rate': 2e-05, 'epoch': 0.63}
+0: 2:46:07,  1.82s/it]
 63%|██████▎   | 9263/14725 [4:43:30<2:44:03,  1.80s/it]
 63%|██████▎   | 9264/14725 [4:43:32<2:45:51,  1.82s/it]
 63%|██████▎   | 9265/14725 [4:43:34<2:44:21,  1.81s/it]
 63%|██████▎   | 9266/14725 [4:43:36<2:42:24,  1.79s/it]
 63%|██████▎   | 9267/14725 [4:43:38<2:44:28,  1.81s/it]
 63%|██████▎   | 9268/14725 [4:43:39<2:42:44,  1.79s/it]
 63%|██████▎   | 9269/14725 [4:43:41<2:41:23,  1.77s/it]
 63%|██████▎   | 9270/14725 [4:43:43<2:42:20,  1.79s/it]
                                                        

 63%|██████▎   | 9270/14725 [4:43:43<2:42:20,  1.79s/it]
 63%|██████▎   | 9271/14725 [4:43:45<2:42:55,  1.79s/it]
 63%|██████▎   | 9272/14725 [4:43:47<2:42:30,  1.79s/it]
 63%|██████▎   | 9273/14725 [4:43:48<2:41:46,  1.78s/it]
 63%|██████▎   | 9274/14725 [4:43:50<2:45:13,  1.82s/it]
 63%|██████
+0: {'loss': 0.8151, 'grad_norm': 0.7084289820174007, 'learning_rate': 2e-05, 'epoch': 0.63}
+0: ▎   | 9275/14725 [4:43:52<2:43:28,  1.80s/it]
 63%|██████▎   | 9276/14725 [4:43:54<2:44:38,  1.81s/it]
 63%|██████▎   | 9277/14725 [4:43:56<2:43:14,  1.80s/it]
 63%|██████▎   | 9278/14725 [4:43:57<2:43:56,  1.81s/it]
 63%|██████▎   | 9279/14725 [4:43:59<2:44:29,  1.81s/it]
 63%|██████▎   | 9280/14725 [4:44:01<2:46:58,  1.84s/it]
                                                        

 63%|██████▎   | 9280/14725 [4:44:01<2:46:58,  1.84s/it]
 63%|██████▎   | 9281/14725 [4:44:03<2:45:42,  1.83s/it]
 63%|██████▎   | 9282/14725 [4:44:05<2:44:21,  1.81s/it]
 63%|██████▎   | 9283/14725 [4:44:06<2:43:22,  1.80s/it]
 63%|██████▎   | 9284/14725 [4:44:08<2:42:03,  1.79s/it]
 63%|██████▎   | 9285/14725 [4:44:10<2:43:13,  1.80s/it]
 63%|██████▎   | 9286/14725 [4:44:12<2:41:53,  1.79s/it]
 63%|██████▎   | 9287/14725 [4:44:14<2:42:21,  1.79s
+0: {'loss': 0.8145, 'grad_norm': 0.7076019452198, 'learning_rate': 2e-05, 'epoch': 0.63}
+0: /it]
 63%|██████▎   | 9288/14725 [4:44:15<2:42:35,  1.79s/it]
 63%|██████▎   | 9289/14725 [4:44:17<2:42:40,  1.80s/it]
 63%|██████▎   | 9290/14725 [4:44:19<2:43:42,  1.81s/it]
                                                        

 63%|██████▎   | 9290/14725 [4:44:19<2:43:42,  1.81s/it]
 63%|██████▎   | 9291/14725 [4:44:21<2:43:33,  1.81s/it]
 63%|██████▎   | 9292/14725 [4:44:23<2:43:14,  1.80s/it]
 63%|██████▎   | 9293/14725 [4:44:24<2:42:20,  1.79s/it]
 63%|██████▎   | 9294/14725 [4:44:26<2:41:41,  1.79s/it]
 63%|██████▎   | 9295/14725 [4:44:28<2:45:01,  1.82s/it]
 63%|██████▎   | 9296/14725 [4:44:30<2:43:29,  1.81s/it]
 63%|██████▎   | 9297/14725 [4:44:32<2:43:10,  1.80s/it]
 63%|██████▎   | 9298/14725 [4:44:34<2:44:07,  1.81s/it]
 63%|██████▎   | 9299/14725 [4:44:35<2:44:55,  1.82s/it]
 63%|██████▎   | 9300/14
+0: {'loss': 0.8091, 'grad_norm': 0.7167013183566019, 'learning_rate': 2e-05, 'epoch': 0.63}
+0: {'loss': 0.8, 'grad_norm': 0.7064826636717696, 'learning_rate': 2e-05, 'epoch': 0.63}
+0: 725 [4:44:37<2:43:40,  1.81s/it]
                                                        

 63%|██████▎   | 9300/14725 [4:44:37<2:43:40,  1.81s/it]
 63%|██████▎   | 9301/14725 [4:44:39<2:43:34,  1.81s/it]
 63%|██████▎   | 9302/14725 [4:44:41<2:42:50,  1.80s/it]
 63%|██████▎   | 9303/14725 [4:44:43<2:42:25,  1.80s/it]
 63%|██████▎   | 9304/14725 [4:44:44<2:42:13,  1.80s/it]
 63%|██████▎   | 9305/14725 [4:44:46<2:41:14,  1.79s/it]
 63%|██████▎   | 9306/14725 [4:44:48<2:40:41,  1.78s/it]
 63%|██████▎   | 9307/14725 [4:44:50<2:42:51,  1.80s/it]
 63%|██████▎   | 9308/14725 [4:44:52<2:43:28,  1.81s/it]
 63%|██████▎   | 9309/14725 [4:44:53<2:44:39,  1.82s/it]
 63%|██████▎   | 9310/14725 [4:44:55<2:43:06,  1.81s/it]
                                                        

 63%|██████▎   | 9310/14725 [4:44:55<2:43:06,  1.81s/it]
 63%|██████
+0: {'loss': 0.8033, 'grad_norm': 0.6607901908544832, 'learning_rate': 2e-05, 'epoch': 0.63}
+0: ▎   | 9311/14725 [4:44:57<2:43:57,  1.82s/it]
 63%|██████▎   | 9312/14725 [4:44:59<2:42:45,  1.80s/it]
 63%|██████▎   | 9313/14725 [4:45:01<2:42:43,  1.80s/it]
 63%|██████▎   | 9314/14725 [4:45:02<2:43:25,  1.81s/it]
 63%|██████▎   | 9315/14725 [4:45:04<2:41:54,  1.80s/it]
 63%|██████▎   | 9316/14725 [4:45:06<2:43:47,  1.82s/it]
 63%|██████▎   | 9317/14725 [4:45:08<2:43:14,  1.81s/it]
 63%|██████▎   | 9318/14725 [4:45:10<2:42:02,  1.80s/it]
 63%|██████▎   | 9319/14725 [4:45:11<2:44:13,  1.82s/it]
 63%|██████▎   | 9320/14725 [4:45:13<2:42:52,  1.81s/it]
                                                        

 63%|██████▎   | 9320/14725 [4:45:13<2:42:52,  1.81s/it]
 63%|██████▎   | 9321/14725 [4:45:15<2:41:41,  1.80s/it]
 63%|██████▎   | 9322/14725 [4:45:17<2:43:08,  1.81s/it]
 63%|██████▎   | 9323/14725 [4:45:19<2:44:12,  1.82s
+0: {'loss': 0.8057, 'grad_norm': 0.6637730972925512, 'learning_rate': 2e-05, 'epoch': 0.63}
+0: /it]
 63%|██████▎   | 9324/14725 [4:45:20<2:42:49,  1.81s/it]
 63%|██████▎   | 9325/14725 [4:45:22<2:42:39,  1.81s/it]
 63%|██████▎   | 9326/14725 [4:45:24<2:41:31,  1.80s/it]
 63%|██████▎   | 9327/14725 [4:45:26<2:43:23,  1.82s/it]
 63%|██████▎   | 9328/14725 [4:45:28<2:43:05,  1.81s/it]
 63%|██████▎   | 9329/14725 [4:45:30<2:42:13,  1.80s/it]
 63%|██████▎   | 9330/14725 [4:45:31<2:43:46,  1.82s/it]
                                                        

 63%|██████▎   | 9330/14725 [4:45:31<2:43:46,  1.82s/it]
 63%|██████▎   | 9331/14725 [4:45:33<2:45:08,  1.84s/it]
 63%|██████▎   | 9332/14725 [4:45:35<2:44:21,  1.83s/it]
 63%|██████▎   | 9333/14725 [4:45:37<2:44:05,  1.83s/it]
 63%|██████▎   | 9334/14725 [4:45:39<2:43:17,  1.82s/it]
 63%|██████▎   | 9335/14725 [4:45:40<2:42:22,  1.81s/it]
 63%|██████▎   | 9336/14
+0: {'loss': 0.8185, 'grad_norm': 0.6235357652456042, 'learning_rate': 2e-05, 'epoch': 0.63}
+0: 725 [4:45:42<2:41:59,  1.80s/it]
 63%|██████▎   | 9337/14725 [4:45:44<2:41:48,  1.80s/it]
 63%|██████▎   | 9338/14725 [4:45:46<2:40:30,  1.79s/it]
 63%|██████▎   | 9339/14725 [4:45:48<2:42:19,  1.81s/it]
 63%|█████��▎   | 9340/14725 [4:45:49<2:40:31,  1.79s/it]
                                                        

 63%|██████▎   | 9340/14725 [4:45:49<2:40:31,  1.79s/it]
 63%|██████▎   | 9341/14725 [4:45:51<2:40:18,  1.79s/it]
 63%|██████▎   | 9342/14725 [4:45:53<2:41:46,  1.80s/it]
 63%|██████▎   | 9343/14725 [4:45:55<2:42:27,  1.81s/it]
 63%|██████▎   | 9344/14725 [4:45:57<2:43:00,  1.82s/it]
 63%|██████▎   | 9345/14725 [4:45:58<2:42:00,  1.81s/it]
 63%|██████▎   | 9346/14725 [4:46:00<2:41:20,  1.80s/it]
 63%|██████▎   | 9347/14725 [4:46:02<2:43:15,  1.82s/it]
 63%|██████▎   | 9348/14725 [4:46:04<2:41:17,  1.80s/it]
 63%|█�
+0: {'loss': 0.8314, 'grad_norm': 0.7386223909055971, 'learning_rate': 2e-05, 'epoch': 0.63}
+0: �████▎   | 9349/14725 [4:46:06<2:40:14,  1.79s/it]
 63%|██████▎   | 9350/14725 [4:46:07<2:41:23,  1.80s/it]
                                                        

 63%|██████▎   | 9350/14725 [4:46:07<2:41:23,  1.80s/it]
 64%|██████▎   | 9351/14725 [4:46:09<2:40:34,  1.79s/it]
 64%|██████▎   | 9352/14725 [4:46:11<2:41:04,  1.80s/it]
 64%|██████▎   | 9353/14725 [4:46:13<2:41:33,  1.80s/it]
 64%|██████▎   | 9354/14725 [4:46:15<2:41:39,  1.81s/it]
 64%|██████▎   | 9355/14725 [4:46:16<2:40:57,  1.80s/it]
 64%|██████▎   | 9356/14725 [4:46:18<2:42:06,  1.81s/it]
 64%|██████▎   | 9357/14725 [4:46:20<2:41:52,  1.81s/it]
 64%|██████▎   | 9358/14725 [4:46:22<2:40:52,  1.80s/it]
 64%|██████▎   | 9359/14725 [4:46:24<2:41:14,  1.80s/it]
 64%|██████▎   | 9360/14725 [4:46:26<2:42:07,  1.81s/it]
                                                     
+0: {'loss': 0.8124, 'grad_norm': 0.7406954084109945, 'learning_rate': 2e-05, 'epoch': 0.64}
+0: {'loss': 0.8269, 'grad_norm': 0.6711939898748637, 'learning_rate': 2e-05, 'epoch': 0.64}
+0:    

 64%|██████▎   | 9360/14725 [4:46:26<2:42:07,  1.81s/it]
 64%|██████▎   | 9361/14725 [4:46:27<2:40:23,  1.79s/it]
 64%|██████▎   | 9362/14725 [4:46:29<2:42:06,  1.81s/it]
 64%|██████▎   | 9363/14725 [4:46:31<2:41:51,  1.81s/it]
 64%|██████▎   | 9364/14725 [4:46:33<2:41:56,  1.81s/it]
 64%|██████▎   | 9365/14725 [4:46:35<2:41:32,  1.81s/it]
 64%|██████▎   | 9366/14725 [4:46:36<2:43:10,  1.83s/it]
 64%|██████▎   | 9367/14725 [4:46:38<2:43:33,  1.83s/it]
 64%|██████▎   | 9368/14725 [4:46:40<2:43:39,  1.83s/it]
 64%|██████▎   | 9369/14725 [4:46:42<2:45:05,  1.85s/it]
 64%|██████▎   | 9370/14725 [4:46:44<2:43:28,  1.83s/it]
                                                        

 64%|██████▎   | 9370/14725 [4:46:44<2:43:28,  1.83s/it]
 64%|██████▎   | 9371/14725 [4:46:46<2:43:58,  1.84s/it]
 64%|██████▎   | 9372/14
+0: {'loss': 0.8115, 'grad_norm': 0.7032491657680956, 'learning_rate': 2e-05, 'epoch': 0.64}
+0: 725 [4:46:47<2:44:14,  1.84s/it]
 64%|██████▎   | 9373/14725 [4:46:49<2:42:24,  1.82s/it]
 64%|██████▎   | 9374/14725 [4:46:51<2:41:10,  1.81s/it]
 64%|██████▎   | 9375/14725 [4:46:53<2:39:47,  1.79s/it]
 64%|██████▎   | 9376/14725 [4:46:55<2:39:18,  1.79s/it]
 64%|██████▎   | 9377/14725 [4:46:56<2:38:03,  1.77s/it]
 64%|██████▎   | 9378/14725 [4:46:58<2:39:04,  1.79s/it]
 64%|██████▎   | 9379/14725 [4:47:00<2:38:58,  1.78s/it]
 64%|██████▎   | 9380/14725 [4:47:02<2:40:19,  1.80s/it]
                                                        

 64%|██████▎   | 9380/14725 [4:47:02<2:40:19,  1.80s/it]
 64%|██████▎   | 9381/14725 [4:47:04<2:42:18,  1.82s/it]
 64%|██████▎   | 9382/14725 [4:47:06<2:44:00,  1.84s/it]
 64%|██████▎   | 9383/14725 [4:47:07<2:42:12,  1.82s/it]
 64%|██████▎   | 9384/14725 [4:47:09<2:41:27,  1.81s/it]
 64%|█�
+0: {'loss': 0.7893, 'grad_norm': 0.6656040704466896, 'learning_rate': 2e-05, 'epoch': 0.64}
+0: �████▎   | 9385/14725 [4:47:11<2:42:27,  1.83s/it]
 64%|██████▎   | 9386/14725 [4:47:13<2:41:20,  1.81s/it]
 64%|██████▎   | 9387/14725 [4:47:15<2:41:47,  1.82s/it]
 64%|██████▍   | 9388/14725 [4:47:16<2:40:58,  1.81s/it]
 64%|██████▍   | 9389/14725 [4:47:18<2:41:01,  1.81s/it]
 64%|██████▍   | 9390/14725 [4:47:20<2:42:23,  1.83s/it]
                                                        

 64%|██████▍   | 9390/14725 [4:47:20<2:42:23,  1.83s/it]
 64%|██████▍   | 9391/14725 [4:47:22<2:40:53,  1.81s/it]
 64%|██████▍   | 9392/14725 [4:47:24<2:39:42,  1.80s/it]
 64%|██████▍   | 9393/14725 [4:47:25<2:41:28,  1.82s/it]
 64%|██████▍   | 9394/14725 [4:47:27<2:41:11,  1.81s/it]
 64%|██████▍   | 9395/14725 [4:47:29<2:40:51,  1.81s/it]
 64%|██████▍   | 9396/14725 [4:47:31<2:40:35,  1.81s/it]
 64%|██████▍   | 9397/14725 [4:47:33<2:
+0: {'loss': 0.8058, 'grad_norm': 0.8795409876055833, 'learning_rate': 2e-05, 'epoch': 0.64}
+0: 39:47,  1.80s/it]
 64%|██████▍   | 9398/14725 [4:47:34<2:38:56,  1.79s/it]
 64%|██████▍   | 9399/14725 [4:47:36<2:38:12,  1.78s/it]
 64%|██████▍   | 9400/14725 [4:47:38<2:41:35,  1.82s/it]
                                                        

 64%|██████▍   | 9400/14725 [4:47:38<2:41:35,  1.82s/it]
 64%|██████▍   | 9401/14725 [4:47:40<2:40:46,  1.81s/it]
 64%|██████▍   | 9402/14725 [4:47:42<2:40:09,  1.81s/it]
 64%|██████▍   | 9403/14725 [4:47:43<2:38:42,  1.79s/it]
 64%|██████▍   | 9404/14725 [4:47:45<2:40:18,  1.81s/it]
 64%|██████▍   | 9405/14725 [4:47:47<2:39:20,  1.80s/it]
 64%|██████▍   | 9406/14725 [4:47:49<2:39:34,  1.80s/it]
 64%|██████▍   | 9407/14725 [4:47:51<2:38:47,  1.79s/it]
 64%|██████▍   | 9408/14725 [4:47:52<2:39:10,  1.80s/it]
 64%|██████▍   | 9409/14725 [4:47:54<2:37:43,  1.78s/it]
 64%|██████�
+0: {'loss': 0.7956, 'grad_norm': 0.7306979111386476, 'learning_rate': 2e-05, 'epoch': 0.64}
+0: {'loss': 0.8019, 'grad_norm': 0.6761645938480043, 'learning_rate': 2e-05, 'epoch': 0.64}
+0: �   | 9410/14725 [4:47:56<2:38:54,  1.79s/it]
                                                        

 64%|██████▍   | 9410/14725 [4:47:56<2:38:54,  1.79s/it]
 64%|██████▍   | 9411/14725 [4:47:58<2:39:07,  1.80s/it]
 64%|██████▍   | 9412/14725 [4:48:00<2:37:57,  1.78s/it]
 64%|██████▍   | 9413/14725 [4:48:01<2:38:54,  1.79s/it]
 64%|██████▍   | 9414/14725 [4:48:03<2:39:43,  1.80s/it]
 64%|██████▍   | 9415/14725 [4:48:05<2:40:58,  1.82s/it]
 64%|██████▍   | 9416/14725 [4:48:07<2:39:47,  1.81s/it]
 64%|██████▍   | 9417/14725 [4:48:09<2:38:42,  1.79s/it]
 64%|██████▍   | 9418/14725 [4:48:10<2:39:34,  1.80s/it]
 64%|██████▍   | 9419/14725 [4:48:12<2:40:12,  1.81s/it]
 64%|██████▍   | 9420/14725 [4:48:14<2:38:36,  1.79s/it]
                                                        

 64%|██████▍   | 9420/14725 [4:48:14<2:38:36,  1.79s/it]
 64%|█�
+0: {'loss': 0.8016, 'grad_norm': 0.7224911367524589, 'learning_rate': 2e-05, 'epoch': 0.64}
+0: �████▍   | 9421/14725 [4:48:16<2:41:28,  1.83s/it]
 64%|██████▍   | 9422/14725 [4:48:18<2:42:53,  1.84s/it]
 64%|██████▍   | 9423/14725 [4:48:20<2:41:20,  1.83s/it]
 64%|██████▍   | 9424/14725 [4:48:21<2:41:47,  1.83s/it]
 64%|██████▍   | 9425/14725 [4:48:23<2:41:06,  1.82s/it]
 64%|██████▍   | 9426/14725 [4:48:25<2:41:13,  1.83s/it]
 64%|██████▍   | 9427/14725 [4:48:27<2:40:53,  1.82s/it]
 64%|██████▍   | 9428/14725 [4:48:29<2:39:03,  1.80s/it]
 64%|██████▍   | 9429/14725 [4:48:30<2:39:30,  1.81s/it]
 64%|██████▍   | 9430/14725 [4:48:32<2:41:12,  1.83s/it]
                                                        

 64%|██████▍   | 9430/14725 [4:48:32<2:41:12,  1.83s/it]
 64%|██████▍   | 9431/14725 [4:48:34<2:40:07,  1.81s/it]
 64%|██████▍   | 9432/14725 [4:48:36<2:40:45,  1.82s/it]
 64%|██████▍   | 9433/14725 [4:48:38<2:
+0: {'loss': 0.7962, 'grad_norm': 0.6741832915244053, 'learning_rate': 2e-05, 'epoch': 0.64}
+0: 46:21,  1.89s/it]
 64%|██████▍   | 9434/14725 [4:48:40<2:45:37,  1.88s/it]
 64%|██████▍   | 9435/14725 [4:48:42<2:42:40,  1.85s/it]
 64%|██████▍   | 9436/14725 [4:48:43<2:41:38,  1.83s/it]
 64%|██████▍   | 9437/14725 [4:48:45<2:42:52,  1.85s/it]
 64%|██████▍   | 9438/14725 [4:48:47<2:40:35,  1.82s/it]
 64%|██████▍   | 9439/14725 [4:48:49<2:39:58,  1.82s/it]
 64%|██████▍   | 9440/14725 [4:48:51<2:40:40,  1.82s/it]
                                                        

 64%|██████▍   | 9440/14725 [4:48:51<2:40:40,  1.82s/it]
 64%|██████▍   | 9441/14725 [4:48:52<2:41:06,  1.83s/it]
 64%|██████▍   | 9442/14725 [4:48:54<2:41:10,  1.83s/it]
 64%|██████▍   | 9443/14725 [4:48:56<2:42:40,  1.85s/it]
 64%|██████▍   | 9444/14725 [4:48:58<2:43:47,  1.86s/it]
 64%|██████▍   | 9445/14725 [4:49:00<2:42:56,  1.85s/it]
 64%|██████�
+0: {'loss': 0.8212, 'grad_norm': 0.7676205137132687, 'learning_rate': 2e-05, 'epoch': 0.64}
+0: �   | 9446/14725 [4:49:02<2:43:06,  1.85s/it]
 64%|██████▍   | 9447/14725 [4:49:04<2:41:22,  1.83s/it]
 64%|██████▍   | 9448/14725 [4:49:05<2:40:50,  1.83s/it]
 64%|██████▍   | 9449/14725 [4:49:07<2:40:39,  1.83s/it]
 64%|██████▍   | 9450/14725 [4:49:09<2:40:20,  1.82s/it]
                                                        

 64%|██████▍   | 9450/14725 [4:49:09<2:40:20,  1.82s/it]
 64%|██████▍   | 9451/14725 [4:49:11<2:39:10,  1.81s/it]
 64%|██████▍   | 9452/14725 [4:49:13<2:38:09,  1.80s/it]
 64%|██████▍   | 9453/14725 [4:49:14<2:38:29,  1.80s/it]
 64%|██████▍   | 9454/14725 [4:49:16<2:38:12,  1.80s/it]
 64%|██████▍   | 9455/14725 [4:49:18<2:39:55,  1.82s/it]
 64%|██████▍   | 9456/14725 [4:49:20<2:40:15,  1.82s/it]
 64%|██████▍   | 9457/14725 [4:49:22<2:40:26,  1.83s/it]
 64%|██████▍   | 9458/14725 [4:49:24<2:40:03,  1.82s/i
+0: {'loss': 0.8224, 'grad_norm': 0.7624618619935322, 'learning_rate': 2e-05, 'epoch': 0.64}
+0: t]
 64%|██████▍   | 9459/14725 [4:49:25<2:38:44,  1.81s/it]
 64%|██████▍   | 9460/14725 [4:49:27<2:39:31,  1.82s/it]
                                                        

 64%|██████▍   | 9460/14725 [4:49:27<2:39:31,  1.82s/it]
 64%|██████▍   | 9461/14725 [4:49:29<2:38:31,  1.81s/it]
 64%|██████▍   | 9462/14725 [4:49:31<2:38:24,  1.81s/it]
 64%|██████▍   | 9463/14725 [4:49:33<2:37:43,  1.80s/it]
 64%|██████▍   | 9464/14725 [4:49:34<2:38:17,  1.81s/it]
 64%|██████▍   | 9465/14725 [4:49:36<2:38:19,  1.81s/it]
 64%|██████▍   | 9466/14725 [4:49:38<2:37:28,  1.80s/it]
 64%|██████▍   | 9467/14725 [4:49:40<2:36:30,  1.79s/it]
 64%|██████▍   | 9468/14725 [4:49:42<2:37:50,  1.80s/it]
 64%|██████▍   | 9469/14725 [4:49:43<2:36:12,  1.78s/it]
 64%|██████▍   | 9470/14725 [4:49:45<2:35:35,  1.78s/it]
                                        
+0: {'loss': 0.805, 'grad_norm': 0.6826810811700572, 'learning_rate': 2e-05, 'epoch': 0.64}
+0: {'loss': 0.8233, 'grad_norm': 0.6866279296533782, 'learning_rate': 2e-05, 'epoch': 0.64}
+0:                 

 64%|██████▍   | 9470/14725 [4:49:45<2:35:35,  1.78s/it]
 64%|██████▍   | 9471/14725 [4:49:47<2:36:42,  1.79s/it]
 64%|██████▍   | 9472/14725 [4:49:49<2:36:47,  1.79s/it]
 64%|██████▍   | 9473/14725 [4:49:50<2:36:04,  1.78s/it]
 64%|██████▍   | 9474/14725 [4:49:52<2:37:48,  1.80s/it]
 64%|██████▍   | 9475/14725 [4:49:54<2:36:11,  1.79s/it]
 64%|██████▍   | 9476/14725 [4:49:56<2:36:54,  1.79s/it]
 64%|██████▍   | 9477/14725 [4:49:58<2:36:17,  1.79s/it]
 64%|██████▍   | 9478/14725 [4:49:59<2:36:28,  1.79s/it]
 64%|██████▍   | 9479/14725 [4:50:01<2:37:20,  1.80s/it]
 64%|██████▍   | 9480/14725 [4:50:03<2:37:35,  1.80s/it]
                                                        

 64%|██████▍   | 9480/14725 [4:50:03<2:37:35,  1.80s/it]
 64%|██████▍   | 9481/14725 [4:50:05<2:44:10,  1.88s/it]
 64%|██████�
+0: {'loss': 0.8139, 'grad_norm': 0.7146102337672309, 'learning_rate': 2e-05, 'epoch': 0.64}
+0: �   | 9482/14725 [4:50:07<2:41:15,  1.85s/it]
 64%|██████▍   | 9483/14725 [4:50:09<2:39:12,  1.82s/it]
 64%|██████▍   | 9484/14725 [4:50:10<2:37:42,  1.81s/it]
 64%|██████▍   | 9485/14725 [4:50:12<2:38:02,  1.81s/it]
 64%|██████▍   | 9486/14725 [4:50:14<2:41:18,  1.85s/it]
 64%|██████▍   | 9487/14725 [4:50:16<2:41:06,  1.85s/it]
 64%|██████▍   | 9488/14725 [4:50:18<2:40:32,  1.84s/it]
 64%|██████▍   | 9489/14725 [4:50:20<2:38:34,  1.82s/it]
 64%|██████▍   | 9490/14725 [4:50:21<2:37:20,  1.80s/it]
                                                        

 64%|██████▍   | 9490/14725 [4:50:21<2:37:20,  1.80s/it]
 64%|██████▍   | 9491/14725 [4:50:23<2:39:22,  1.83s/it]
 64%|██████▍   | 9492/14725 [4:50:25<2:38:20,  1.82s/it]
 64%|██████▍   | 9493/14725 [4:50:27<2:38:32,  1.82s/it]
 64%|██████▍   | 9494/14725 [4:50:29<2:36:40,  1.80s/i
+0: {'loss': 0.7947, 'grad_norm': 0.6577960715139299, 'learning_rate': 2e-05, 'epoch': 0.65}
+0: t]
 64%|██████▍   | 9495/14725 [4:50:30<2:37:02,  1.80s/it]
 64%|██████▍   | 9496/14725 [4:50:32<2:36:19,  1.79s/it]
 64%|██████▍   | 9497/14725 [4:50:34<2:37:34,  1.81s/it]
 65%|██████▍   | 9498/14725 [4:50:36<2:38:21,  1.82s/it]
 65%|██████▍   | 9499/14725 [4:50:38<2:38:05,  1.82s/it]
 65%|██████▍   | 9500/14725 [4:50:39<2:36:37,  1.80s/it]
                                                        

 65%|██████▍   | 9500/14725 [4:50:39<2:36:37,  1.80s/it]
 65%|██████▍   | 9501/14725 [4:50:41<2:37:11,  1.81s/it]
 65%|██████▍   | 9502/14725 [4:50:43<2:36:27,  1.80s/it]
 65%|██████▍   | 9503/14725 [4:50:45<2:36:38,  1.80s/it]
 65%|██████▍   | 9504/14725 [4:50:47<2:39:17,  1.83s/it]
 65%|██████▍   | 9505/14725 [4:50:49<2:39:07,  1.83s/it]
 65%|██████▍   | 9506/14725 [4:50:50<2:39:56,  1.84s/it]
 65%|██████▍   | 9507/1472
+0: {'loss': 0.8105, 'grad_norm': 0.7149476966749704, 'learning_rate': 2e-05, 'epoch': 0.65}
+0: 5 [4:50:52<2:41:01,  1.85s/it]
 65%|██████▍   | 9508/14725 [4:50:54<2:39:12,  1.83s/it]
 65%|██████▍   | 9509/14725 [4:50:56<2:38:19,  1.82s/it]
 65%|██████▍   | 9510/14725 [4:50:58<2:36:36,  1.80s/it]
                                                        

 65%|██████▍   | 9510/14725 [4:50:58<2:36:36,  1.80s/it]
 65%|██████▍   | 9511/14725 [4:51:00<2:39:03,  1.83s/it]
 65%|██████▍   | 9512/14725 [4:51:01<2:37:57,  1.82s/it]
 65%|██████▍   | 9513/14725 [4:51:03<2:38:18,  1.82s/it]
 65%|██████▍   | 9514/14725 [4:51:05<2:37:27,  1.81s/it]
 65%|██████▍   | 9515/14725 [4:51:07<2:38:29,  1.83s/it]
 65%|██████▍   | 9516/14725 [4:51:09<2:39:51,  1.84s/it]
 65%|██████▍   | 9517/14725 [4:51:10<2:39:13,  1.83s/it]
 65%|██████▍   | 9518/14725 [4:51:12<2:37:23,  1.81s/it]
 65%|██████▍   | 9519/14725 [4:51:14<2:36:54,  1.81s/it]
 65%|██�
+0: {'loss': 0.8332, 'grad_norm': 0.7495108326740779, 'learning_rate': 2e-05, 'epoch': 0.65}
+0: {'loss': 0.8027, 'grad_norm': 0.6992291126079548, 'learning_rate': 2e-05, 'epoch': 0.65}
+0: ��███▍   | 9520/14725 [4:51:16<2:37:11,  1.81s/it]
                                                        

 65%|██████▍   | 9520/14725 [4:51:16<2:37:11,  1.81s/it]
 65%|██████▍   | 9521/14725 [4:51:18<2:36:27,  1.80s/it]
 65%|██████▍   | 9522/14725 [4:51:19<2:36:56,  1.81s/it]
 65%|██████▍   | 9523/14725 [4:51:21<2:37:56,  1.82s/it]
 65%|██████▍   | 9524/14725 [4:51:23<2:37:17,  1.81s/it]
 65%|██████▍   | 9525/14725 [4:51:25<2:36:03,  1.80s/it]
 65%|██████▍   | 9526/14725 [4:51:27<2:35:26,  1.79s/it]
 65%|██████▍   | 9527/14725 [4:51:29<2:46:53,  1.93s/it]
 65%|██████▍   | 9528/14725 [4:51:31<2:42:42,  1.88s/it]
 65%|██████▍   | 9529/14725 [4:51:32<2:38:52,  1.83s/it]
 65%|██████▍   | 9530/14725 [4:51:34<2:36:34,  1.81s/it]
                                                        

 65%|██████▍   | 9530/14725 [4:51:34<2:36:34,  1.81s/i
+0: {'loss': 0.8165, 'grad_norm': 0.7147090160889414, 'learning_rate': 2e-05, 'epoch': 0.65}
+0: t]
 65%|██████▍   | 9531/14725 [4:51:36<2:37:53,  1.82s/it]
 65%|██████▍   | 9532/14725 [4:51:38<2:36:05,  1.80s/it]
 65%|██████▍   | 9533/14725 [4:51:40<2:35:43,  1.80s/it]
 65%|██████▍   | 9534/14725 [4:51:41<2:35:57,  1.80s/it]
 65%|██████▍   | 9535/14725 [4:51:43<2:34:29,  1.79s/it]
 65%|██████▍   | 9536/14725 [4:51:45<2:34:01,  1.78s/it]
 65%|██████▍   | 9537/14725 [4:51:47<2:35:34,  1.80s/it]
 65%|██████▍   | 9538/14725 [4:51:49<2:36:04,  1.81s/it]
 65%|██████▍   | 9539/14725 [4:51:50<2:35:56,  1.80s/it]
 65%|██████▍   | 9540/14725 [4:51:52<2:34:20,  1.79s/it]
                                                        

 65%|██████▍   | 9540/14725 [4:51:52<2:34:20,  1.79s/it]
 65%|██████▍   | 9541/14725 [4:51:54<2:36:00,  1.81s/it]
 65%|██████▍   | 9542/14725 [4:51:56<2:35:09,  1.80s/it]
 65%|██████▍   | 9543/1472
+0: {'loss': 0.8245, 'grad_norm': 0.6975941020290172, 'learning_rate': 2e-05, 'epoch': 0.65}
+0: 5 [4:51:57<2:33:45,  1.78s/it]
 65%|██████▍   | 9544/14725 [4:51:59<2:34:44,  1.79s/it]
 65%|██████▍   | 9545/14725 [4:52:01<2:37:20,  1.82s/it]
 65%|██████▍   | 9546/14725 [4:52:03<2:35:20,  1.80s/it]
 65%|██████▍   | 9547/14725 [4:52:05<2:42:27,  1.88s/it]
 65%|██████▍   | 9548/14725 [4:52:07<2:40:40,  1.86s/it]
 65%|██████▍   | 9549/14725 [4:52:09<2:41:44,  1.87s/it]
 65%|██████▍   | 9550/14725 [4:52:11<2:41:16,  1.87s/it]
                                                        

 65%|██████▍   | 9550/14725 [4:52:11<2:41:16,  1.87s/it]
 65%|██████▍   | 9551/14725 [4:52:12<2:39:40,  1.85s/it]
 65%|██████▍   | 9552/14725 [4:52:14<2:38:05,  1.83s/it]
 65%|██████▍   | 9553/14725 [4:52:16<2:36:59,  1.82s/it]
 65%|██████▍   | 9554/14725 [4:52:18<2:38:25,  1.84s/it]
 65%|██████▍   | 9555/14725 [4:52:20<2:37:15,  1.83s/it]
 65%|██�
+0: {'loss': 0.7967, 'grad_norm': 0.6942130360123188, 'learning_rate': 2e-05, 'epoch': 0.65}
+0: ��███▍   | 9556/14725 [4:52:21<2:36:17,  1.81s/it]
 65%|██████▍   | 9557/14725 [4:52:23<2:36:26,  1.82s/it]
 65%|██████▍   | 9558/14725 [4:52:25<2:35:22,  1.80s/it]
 65%|██████▍   | 9559/14725 [4:52:27<2:34:41,  1.80s/it]
 65%|██████▍   | 9560/14725 [4:52:29<2:42:51,  1.89s/it]
                                                        

 65%|██████▍   | 9560/14725 [4:52:29<2:42:51,  1.89s/it]
 65%|██████▍   | 9561/14725 [4:52:31<2:40:02,  1.86s/it]
 65%|██████▍   | 9562/14725 [4:52:32<2:37:15,  1.83s/it]
 65%|██████▍   | 9563/14725 [4:52:34<2:35:24,  1.81s/it]
 65%|██████▍   | 9564/14725 [4:52:36<2:35:21,  1.81s/it]
 65%|██████▍   | 9565/14725 [4:52:38<2:35:28,  1.81s/it]
 65%|██████▍   | 9566/14725 [4:52:40<2:37:35,  1.83s/it]
 65%|██████▍   | 9567/14725 [4:52:42<2:39:40,  1.86s/it]
 65%|██████▍   | 9568/14725 [4:52:43<2:37
+0: {'loss': 0.8098, 'grad_norm': 0.659858285949057, 'learning_rate': 2e-05, 'epoch': 0.65}
+0: :24,  1.83s/it]
 65%|██████▍   | 9569/14725 [4:52:45<2:36:01,  1.82s/it]
 65%|██████▍   | 9570/14725 [4:52:47<2:34:49,  1.80s/it]
                                                        

 65%|██████▍   | 9570/14725 [4:52:47<2:34:49,  1.80s/it]
 65%|██████▍   | 9571/14725 [4:52:49<2:35:22,  1.81s/it]
 65%|██████▌   | 9572/14725 [4:52:51<2:34:19,  1.80s/it]
 65%|██████▌   | 9573/14725 [4:52:52<2:33:30,  1.79s/it]
 65%|██████▌   | 9574/14725 [4:52:54<2:33:22,  1.79s/it]
 65%|██████▌   | 9575/14725 [4:52:56<2:32:10,  1.77s/it]
 65%|██████▌   | 9576/14725 [4:52:58<2:32:09,  1.77s/it]
 65%|██████▌   | 9577/14725 [4:52:59<2:32:24,  1.78s/it]
 65%|██████▌   | 9578/14725 [4:53:01<2:32:09,  1.77s/it]
 65%|██████▌   | 9579/14725 [4:53:03<2:36:33,  1.83s/it]
 65%|██████▌   | 9580/14725 [4:53:05<2:34:50,  1.81s/it]
                           
+0: {'loss': 0.8141, 'grad_norm': 0.677376107633859, 'learning_rate': 2e-05, 'epoch': 0.65}
+0: {'loss': 0.8024, 'grad_norm': 0.6780311210503637, 'learning_rate': 2e-05, 'epoch': 0.65}
+0:                              

 65%|██████▌   | 9580/14725 [4:53:05<2:34:50,  1.81s/it]
 65%|██████▌   | 9581/14725 [4:53:07<2:42:46,  1.90s/it]
 65%|██████▌   | 9582/14725 [4:53:09<2:39:19,  1.86s/it]
 65%|██████▌   | 9583/14725 [4:53:11<2:36:40,  1.83s/it]
 65%|██████▌   | 9584/14725 [4:53:12<2:36:20,  1.82s/it]
 65%|██████▌   | 9585/14725 [4:53:14<2:34:57,  1.81s/it]
 65%|██████▌   | 9586/14725 [4:53:16<2:35:47,  1.82s/it]
 65%|██████▌   | 9587/14725 [4:53:18<2:34:13,  1.80s/it]
 65%|██████▌   | 9588/14725 [4:53:20<2:34:18,  1.80s/it]
 65%|██████▌   | 9589/14725 [4:53:21<2:36:12,  1.82s/it]
 65%|██████▌   | 9590/14725 [4:53:23<2:36:05,  1.82s/it]
                                                        

 65%|██████▌   | 9590/14725 [4:53:23<2:36:05,  1.82s/it]
 65%|██████▌   | 9591/14725 [4:53:25<2:35:03,  1.81s/it]
 65%|██�
+0: {'loss': 0.7906, 'grad_norm': 0.7237530950906956, 'learning_rate': 2e-05, 'epoch': 0.65}
+0: ��███▌   | 9592/14725 [4:53:27<2:35:44,  1.82s/it]
 65%|██████▌   | 9593/14725 [4:53:29<2:36:36,  1.83s/it]
 65%|██████▌   | 9594/14725 [4:53:31<2:40:41,  1.88s/it]
 65%|██████▌   | 9595/14725 [4:53:32<2:38:03,  1.85s/it]
 65%|██████▌   | 9596/14725 [4:53:34<2:36:03,  1.83s/it]
 65%|██████▌   | 9597/14725 [4:53:36<2:37:46,  1.85s/it]
 65%|██████▌   | 9598/14725 [4:53:38<2:35:53,  1.82s/it]
 65%|██████▌   | 9599/14725 [4:53:40<2:35:04,  1.82s/it]
 65%|██████▌   | 9600/14725 [4:53:42<2:36:04,  1.83s/it]
                                                        

 65%|██████▌   | 9600/14725 [4:53:42<2:36:04,  1.83s/it]
 65%|██████▌   | 9601/14725 [4:53:43<2:37:10,  1.84s/it]
 65%|██████▌   | 9602/14725 [4:53:45<2:36:15,  1.83s/it]
 65%|██████▌   | 9603/14725 [4:53:47<2:34:23,  1.81s/it]
 65%|██████▌   | 9604/14725 [4:53:49<2:35
+0: {'loss': 0.7953, 'grad_norm': 0.6930308169913931, 'learning_rate': 2e-05, 'epoch': 0.65}
+0: :46,  1.83s/it]
 65%|██████▌   | 9605/14725 [4:53:51<2:34:33,  1.81s/it]
 65%|██████▌   | 9606/14725 [4:53:52<2:33:58,  1.80s/it]
 65%|██████▌   | 9607/14725 [4:53:54<2:32:59,  1.79s/it]
 65%|██████▌   | 9608/14725 [4:53:56<2:31:54,  1.78s/it]
 65%|██████▌   | 9609/14725 [4:53:58<2:33:52,  1.80s/it]
 65%|██████▌   | 9610/14725 [4:54:00<2:33:35,  1.80s/it]
                                                        

 65%|██████▌   | 9610/14725 [4:54:00<2:33:35,  1.80s/it]
 65%|██████▌   | 9611/14725 [4:54:01<2:32:58,  1.79s/it]
 65%|██████▌   | 9612/14725 [4:54:03<2:35:14,  1.82s/it]
 65%|█���████▌   | 9613/14725 [4:54:05<2:33:46,  1.80s/it]
 65%|██████▌   | 9614/14725 [4:54:07<2:33:17,  1.80s/it]
 65%|██████▌   | 9615/14725 [4:54:09<2:32:40,  1.79s/it]
 65%|██████▌   | 9616/14725 [4:54:10<2:32:12,  1.79s/it]
 65%|██████▌ 
+0: {'loss': 0.7891, 'grad_norm': 0.683496255565122, 'learning_rate': 2e-05, 'epoch': 0.65}
+0:   | 9617/14725 [4:54:12<2:32:47,  1.79s/it]
 65%|██████▌   | 9618/14725 [4:54:14<2:32:46,  1.79s/it]
 65%|██████▌   | 9619/14725 [4:54:16<2:33:12,  1.80s/it]
 65%|██████▌   | 9620/14725 [4:54:18<2:31:57,  1.79s/it]
                                                        

 65%|██████▌   | 9620/14725 [4:54:18<2:31:57,  1.79s/it]
 65%|██████▌   | 9621/14725 [4:54:19<2:33:55,  1.81s/it]
 65%|██████▌   | 9622/14725 [4:54:21<2:34:53,  1.82s/it]
 65%|██████▌   | 9623/14725 [4:54:23<2:33:41,  1.81s/it]
 65%|██████▌   | 9624/14725 [4:54:25<2:33:39,  1.81s/it]
 65%|██████▌   | 9625/14725 [4:54:27<2:32:12,  1.79s/it]
 65%|██████▌   | 9626/14725 [4:54:28<2:34:44,  1.82s/it]
 65%|██████▌   | 9627/14725 [4:54:30<2:35:10,  1.83s/it]
 65%|██████▌   | 9628/14725 [4:54:32<2:33:33,  1.81s/it]
 65%|██████▌   | 9629/14725 [4:54:34<2:34:16,  1.82s/it]
+0: {'loss': 0.8049, 'grad_norm': 0.7609711049152508, 'learning_rate': 2e-05, 'epoch': 0.65}
+0: {'loss': 0.8126, 'grad_norm': 0.7443533124532696, 'learning_rate': 2e-05, 'epoch': 0.65}
+0: 
 65%|██████▌   | 9630/14725 [4:54:36<2:34:27,  1.82s/it]
                                                        

 65%|██████▌   | 9630/14725 [4:54:36<2:34:27,  1.82s/it]
 65%|██████▌   | 9631/14725 [4:54:38<2:35:13,  1.83s/it]
 65%|██████▌   | 9632/14725 [4:54:39<2:36:25,  1.84s/it]
 65%|██████▌   | 9633/14725 [4:54:41<2:34:24,  1.82s/it]
 65%|██████▌   | 9634/14725 [4:54:43<2:34:04,  1.82s/it]
 65%|██████▌   | 9635/14725 [4:54:45<2:32:32,  1.80s/it]
 65%|██████▌   | 9636/14725 [4:54:47<2:32:27,  1.80s/it]
 65%|██████▌   | 9637/14725 [4:54:49<2:36:29,  1.85s/it]
 65%|██████▌   | 9638/14725 [4:54:50<2:34:36,  1.82s/it]
 65%|██████▌   | 9639/14725 [4:54:52<2:34:18,  1.82s/it]
 65%|██████▌   | 9640/14725 [4:54:54<2:34:04,  1.82s/it]
                                                        

 65%|██████▌   | 9640/14725 [4:54:54<2:34
+0: {'loss': 0.8124, 'grad_norm': 0.6794439175495084, 'learning_rate': 2e-05, 'epoch': 0.66}
+0: :04,  1.82s/it]
 65%|██████▌   | 9641/14725 [4:54:56<2:34:01,  1.82s/it]
 65%|██████▌   | 9642/14725 [4:54:58<2:33:27,  1.81s/it]
 65%|██████▌   | 9643/14725 [4:54:59<2:32:25,  1.80s/it]
 65%|██████▌   | 9644/14725 [4:55:01<2:32:32,  1.80s/it]
 66%|██████▌   | 9645/14725 [4:55:03<2:31:22,  1.79s/it]
 66%|██████▌   | 9646/14725 [4:55:05<2:31:10,  1.79s/it]
 66%|██████▌   | 9647/14725 [4:55:07<2:33:37,  1.82s/it]
 66%|██████▌   | 9648/14725 [4:55:08<2:33:32,  1.81s/it]
 66%|██████▌   | 9649/14725 [4:55:10<2:34:48,  1.83s/it]
 66%|██████▌   | 9650/14725 [4:55:12<2:33:27,  1.81s/it]
                                                        

 66%|██████▌   | 9650/14725 [4:55:12<2:33:27,  1.81s/it]
 66%|██████▌   | 9651/14725 [4:55:14<2:33:24,  1.81s/it]
 66%|██████▌   | 9652/14725 [4:55:16<2:33:10,  1.81s/it]
 66%|██████▌ 
+0: {'loss': 0.8253, 'grad_norm': 0.7095545685980926, 'learning_rate': 2e-05, 'epoch': 0.66}
+0:   | 9653/14725 [4:55:17<2:31:55,  1.80s/it]
 66%|██████▌   | 9654/14725 [4:55:19<2:32:15,  1.80s/it]
 66%|██████▌   | 9655/14725 [4:55:21<2:30:57,  1.79s/it]
 66%|██████▌   | 9656/14725 [4:55:23<2:30:08,  1.78s/it]
 66%|██████▌   | 9657/14725 [4:55:25<2:30:57,  1.79s/it]
 66%|██████▌   | 9658/14725 [4:55:26<2:30:30,  1.78s/it]
 66%|██████▌   | 9659/14725 [4:55:28<2:31:31,  1.79s/it]
 66%|██████▌   | 9660/14725 [4:55:30<2:35:40,  1.84s/it]
                                                        

 66%|██████▌   | 9660/14725 [4:55:30<2:35:40,  1.84s/it]
 66%|██████▌   | 9661/14725 [4:55:32<2:33:55,  1.82s/it]
 66%|██████▌   | 9662/14725 [4:55:34<2:32:30,  1.81s/it]
 66%|██████▌   | 9663/14725 [4:55:35<2:31:45,  1.80s/it]
 66%|██████▌   | 9664/14725 [4:55:37<2:31:20,  1.79s/it]
 66%|██████▌   | 9665/14725 [4:55:39<2:30:41,  1.79s/it]
+0: {'loss': 0.8043, 'grad_norm': 0.7375689147368314, 'learning_rate': 2e-05, 'epoch': 0.66}
+0: 
 66%|██████▌   | 9666/14725 [4:55:41<2:32:09,  1.80s/it]
 66%|██████▌   | 9667/14725 [4:55:43<2:32:33,  1.81s/it]
 66%|██████▌   | 9668/14725 [4:55:45<2:35:08,  1.84s/it]
 66%|██████▌   | 9669/14725 [4:55:46<2:33:32,  1.82s/it]
 66%|██████▌   | 9670/14725 [4:55:48<2:32:33,  1.81s/it]
                                                        

 66%|██████▌   | 9670/14725 [4:55:48<2:32:33,  1.81s/it]
 66%|██████▌   | 9671/14725 [4:55:50<2:33:08,  1.82s/it]
 66%|██████▌   | 9672/14725 [4:55:52<2:39:01,  1.89s/it]
 66%|██████▌   | 9673/14725 [4:55:54<2:35:59,  1.85s/it]
 66%|██████▌   | 9674/14725 [4:55:56<2:34:47,  1.84s/it]
 66%|██████▌   | 9675/14725 [4:55:57<2:34:28,  1.84s/it]
 66%|██████▌   | 9676/14725 [4:55:59<2:34:53,  1.84s/it]
 66%|██████▌   | 9677/14725 [4:56:01<2:33:22,  1.82s/it]
 66%|██████▌   | 9678/14725 
+0: {'loss': 0.8144, 'grad_norm': 0.6586770937068092, 'learning_rate': 2e-05, 'epoch': 0.66}
+0: [4:56:03<2:32:03,  1.81s/it]
 66%|██████▌   | 9679/14725 [4:56:05<2:32:47,  1.82s/it]
 66%|██████▌   | 9680/14725 [4:56:06<2:31:09,  1.80s/it]
                                                        

 66%|██████▌   | 9680/14725 [4:56:06<2:31:09,  1.80s/it]
 66%|██████▌   | 9681/14725 [4:56:08<2:32:18,  1.81s/it]
 66%|██████▌   | 9682/14725 [4:56:10<2:31:25,  1.80s/it]
 66%|██████▌   | 9683/14725 [4:56:12<2:30:31,  1.79s/it]
 66%|██████▌   | 9684/14725 [4:56:14<2:30:35,  1.79s/it]
 66%|██████▌   | 9685/14725 [4:56:15<2:29:17,  1.78s/it]
 66%|██████▌   | 9686/14725 [4:56:17<2:30:19,  1.79s/it]
 66%|██████▌   | 9687/14725 [4:56:19<2:30:04,  1.79s/it]
 66%|██████▌   | 9688/14725 [4:56:21<2:30:00,  1.79s/it]
 66%|██████▌   | 9689/14725 [4:56:22<2:29:32,  1.78s/it]
 66%|██████▌   | 9690/14725 [4:56:24<2:28:48,  1.77s/it]
              
+0: {'loss': 0.7903, 'grad_norm': 0.6718443774385817, 'learning_rate': 2e-05, 'epoch': 0.66}
+0: {'loss': 0.8064, 'grad_norm': 0.71945851698512, 'learning_rate': 2e-05, 'epoch': 0.66}
+0:                                           

 66%|██████▌   | 9690/14725 [4:56:24<2:28:48,  1.77s/it]
 66%|██████▌   | 9691/14725 [4:56:26<2:29:03,  1.78s/it]
 66%|██████▌   | 9692/14725 [4:56:28<2:28:38,  1.77s/it]
 66%|██████▌   | 9693/14725 [4:56:30<2:30:35,  1.80s/it]
 66%|██████▌   | 9694/14725 [4:56:31<2:30:28,  1.79s/it]
 66%|██████▌   | 9695/14725 [4:56:33<2:31:06,  1.80s/it]
 66%|██████▌   | 9696/14725 [4:56:35<2:31:06,  1.80s/it]
 66%|██████▌   | 9697/14725 [4:56:37<2:31:26,  1.81s/it]
 66%|██████▌   | 9698/14725 [4:56:39<2:30:22,  1.79s/it]
 66%|██████▌   | 9699/14725 [4:56:40<2:32:25,  1.82s/it]
 66%|██████▌   | 9700/14725 [4:56:42<2:33:04,  1.83s/it]
                                                        

 66%|██████▌   | 9700/14725 [4:56:42<2:33:04,  1.83s/it]
 66%|██████▌   | 9701/14725 [4:56:44<2:32:38,  1.82s/it]
+0: {'loss': 0.8216, 'grad_norm': 0.7147224240112064, 'learning_rate': 2e-05, 'epoch': 0.66}
+0: 
 66%|██████▌   | 9702/14725 [4:56:46<2:32:37,  1.82s/it]
 66%|██████▌   | 9703/14725 [4:56:48<2:32:56,  1.83s/it]
 66%|██████▌   | 9704/14725 [4:56:50<2:30:59,  1.80s/it]
 66%|██████▌   | 9705/14725 [4:56:51<2:31:49,  1.81s/it]
 66%|██████▌   | 9706/14725 [4:56:53<2:32:37,  1.82s/it]
 66%|██████▌   | 9707/14725 [4:56:55<2:32:08,  1.82s/it]
 66%|██████▌   | 9708/14725 [4:56:57<2:31:13,  1.81s/it]
 66%|██████▌   | 9709/14725 [4:56:59<2:30:12,  1.80s/it]
 66%|██████▌   | 9710/14725 [4:57:00<2:30:57,  1.81s/it]
                                                        

 66%|██████▌   | 9710/14725 [4:57:00<2:30:57,  1.81s/it]
 66%|██████▌   | 9711/14725 [4:57:02<2:30:07,  1.80s/it]
 66%|██████▌   | 9712/14725 [4:57:04<2:33:25,  1.84s/it]
 66%|██████▌   | 9713/14725 [4:57:06<2:31:50,  1.82s/it]
 66%|██████▌   | 9714/14725 
+0: {'loss': 0.8089, 'grad_norm': 0.6568575845121276, 'learning_rate': 2e-05, 'epoch': 0.66}
+0: [4:57:08<2:32:30,  1.83s/it]
 66%|██████▌   | 9715/14725 [4:57:10<2:32:55,  1.83s/it]
 66%|██████▌   | 9716/14725 [4:57:11<2:31:45,  1.82s/it]
 66%|██████▌   | 9717/14725 [4:57:13<2:30:36,  1.80s/it]
 66%|██████▌   | 9718/14725 [4:57:15<2:31:26,  1.81s/it]
 66%|██████▌   | 9719/14725 [4:57:17<2:29:48,  1.80s/it]
 66%|██████▌   | 9720/14725 [4:57:19<2:30:30,  1.80s/it]
                                                        

 66%|██████▌   | 9720/14725 [4:57:19<2:30:30,  1.80s/it]
 66%|██████▌   | 9721/14725 [4:57:20<2:32:21,  1.83s/it]
 66%|██████▌   | 9722/14725 [4:57:22<2:31:04,  1.81s/it]
 66%|██████▌   | 9723/14725 [4:57:24<2:30:37,  1.81s/it]
 66%|██████▌   | 9724/14725 [4:57:26<2:29:48,  1.80s/it]
 66%|██████▌   | 9725/14725 [4:57:28<2:29:27,  1.79s/it]
 66%|██████▌   | 9726/14725 [4:57:29<2:30:08,  1.80s/it]
 66%|███
+0: {'loss': 0.8287, 'grad_norm': 0.715794910184952, 'learning_rate': 2e-05, 'epoch': 0.66}
+0: ███▌   | 9727/14725 [4:57:31<2:29:22,  1.79s/it]
 66%|██████▌   | 9728/14725 [4:57:33<2:30:34,  1.81s/it]
 66%|██████▌   | 9729/14725 [4:57:35<2:29:11,  1.79s/it]
 66%|██████▌   | 9730/14725 [4:57:37<2:32:38,  1.83s/it]
                                                        

 66%|██████▌   | 9730/14725 [4:57:37<2:32:38,  1.83s/it]
 66%|██████▌   | 9731/14725 [4:57:39<2:33:19,  1.84s/it]
 66%|██████▌   | 9732/14725 [4:57:40<2:33:11,  1.84s/it]
 66%|██████▌   | 9733/14725 [4:57:42<2:31:13,  1.82s/it]
 66%|██████▌   | 9734/14725 [4:57:44<2:29:59,  1.80s/it]
 66%|██████▌   | 9735/14725 [4:57:46<2:29:31,  1.80s/it]
 66%|██████▌   | 9736/14725 [4:57:47<2:28:27,  1.79s/it]
 66%|██████▌   | 9737/14725 [4:57:49<2:28:41,  1.79s/it]
 66%|██████▌   | 9738/14725 [4:57:51<2:33:22,  1.85s/it]
 66%|██████▌   | 9739/14725 [4:57:53<2:32:2
+0: {'loss': 0.8015, 'grad_norm': 0.6985880183525934, 'learning_rate': 2e-05, 'epoch': 0.66}
+0: {'loss': 0.8086, 'grad_norm': 0.6763286262684209, 'learning_rate': 2e-05, 'epoch': 0.66}
+0: 4,  1.83s/it]
 66%|██████▌   | 9740/14725 [4:57:55<2:32:50,  1.84s/it]
                                                        

 66%|██████▌   | 9740/14725 [4:57:55<2:32:50,  1.84s/it]
 66%|██████▌   | 9741/14725 [4:57:57<2:31:22,  1.82s/it]
 66%|██████▌   | 9742/14725 [4:57:59<2:31:28,  1.82s/it]
 66%|██████▌   | 9743/14725 [4:58:00<2:30:00,  1.81s/it]
 66%|██████▌   | 9744/14725 [4:58:02<2:32:34,  1.84s/it]
 66%|██████▌   | 9745/14725 [4:58:04<2:30:53,  1.82s/it]
 66%|██████▌   | 9746/14725 [4:58:06<2:29:30,  1.80s/it]
 66%|██████▌   | 9747/14725 [4:58:08<2:29:31,  1.80s/it]
 66%|██████▌   | 9748/14725 [4:58:09<2:28:50,  1.79s/it]
 66%|██████▌   | 9749/14725 [4:58:11<2:29:13,  1.80s/it]
 66%|██████▌   | 9750/14725 [4:58:13<2:28:34,  1.79s/it]
                                                        

 66%|██████▌   | 9750/14725 
+0: {'loss': 0.8043, 'grad_norm': 0.6994578574822177, 'learning_rate': 2e-05, 'epoch': 0.66}
+0: [4:58:13<2:28:34,  1.79s/it]
 66%|██████▌   | 9751/14725 [4:58:15<2:30:08,  1.81s/it]
 66%|██████▌   | 9752/14725 [4:58:17<2:30:33,  1.82s/it]
 66%|██████▌   | 9753/14725 [4:58:18<2:30:19,  1.81s/it]
 66%|██████▌   | 9754/14725 [4:58:20<2:30:56,  1.82s/it]
 66%|██████▌   | 9755/14725 [4:58:22<2:29:57,  1.81s/it]
 66%|██████▋   | 9756/14725 [4:58:24<2:29:28,  1.80s/it]
 66%|██████▋   | 9757/14725 [4:58:26<2:28:57,  1.80s/it]
 66%|██████▋   | 9758/14725 [4:58:28<2:32:11,  1.84s/it]
 66%|██████▋   | 9759/14725 [4:58:29<2:30:17,  1.82s/it]
 66%|██████▋   | 9760/14725 [4:58:31<2:30:30,  1.82s/it]
                                                        

 66%|██████▋   | 9760/14725 [4:58:31<2:30:30,  1.82s/it]
 66%|██████▋   | 9761/14725 [4:58:33<2:30:45,  1.82s/it]
 66%|██████▋   | 9762/14725 [4:58:35<2:28:36,  1.80s/it]
 66%|███
+0: {'loss': 0.8139, 'grad_norm': 0.7098582662992732, 'learning_rate': 2e-05, 'epoch': 0.66}
+0: ███▋   | 9763/14725 [4:58:36<2:27:27,  1.78s/it]
 66%|██████▋   | 9764/14725 [4:58:38<2:27:34,  1.78s/it]
 66%|██████▋   | 9765/14725 [4:58:40<2:26:45,  1.78s/it]
 66%|██████▋   | 9766/14725 [4:58:42<2:28:26,  1.80s/it]
 66%|██████▋   | 9767/14725 [4:58:44<2:28:11,  1.79s/it]
 66%|██████▋   | 9768/14725 [4:58:45<2:28:52,  1.80s/it]
 66%|██████▋   | 9769/14725 [4:58:47<2:30:04,  1.82s/it]
 66%|██████▋   | 9770/14725 [4:58:49<2:32:56,  1.85s/it]
                                                        

 66%|██████▋   | 9770/14725 [4:58:49<2:32:56,  1.85s/it]
 66%|██████▋   | 9771/14725 [4:58:51<2:31:04,  1.83s/it]
 66%|██████▋   | 9772/14725 [4:58:53<2:32:08,  1.84s/it]
 66%|██████▋   | 9773/14725 [4:58:55<2:30:53,  1.83s/it]
 66%|██████▋   | 9774/14725 [4:58:57<2:31:15,  1.83s/it]
 66%|██████▋   | 9775/14725 [4:58:58<2:31:3
+0: {'loss': 0.8096, 'grad_norm': 0.7048679165548016, 'learning_rate': 2e-05, 'epoch': 0.66}
+0: 5,  1.84s/it]
 66%|██████▋   | 9776/14725 [4:59:00<2:31:05,  1.83s/it]
 66%|██████▋   | 9777/14725 [4:59:02<2:29:53,  1.82s/it]
 66%|██████▋   | 9778/14725 [4:59:04<2:33:09,  1.86s/it]
 66%|██████▋   | 9779/14725 [4:59:06<2:30:47,  1.83s/it]
 66%|██████▋   | 9780/14725 [4:59:08<2:34:04,  1.87s/it]
                                                        

 66%|██████▋   | 9780/14725 [4:59:08<2:34:04,  1.87s/it]
 66%|██████▋   | 9781/14725 [4:59:09<2:32:19,  1.85s/it]
 66%|██████▋   | 9782/14725 [4:59:11<2:29:35,  1.82s/it]
 66%|██████▋   | 9783/14725 [4:59:13<2:29:43,  1.82s/it]
 66%|██████▋   | 9784/14725 [4:59:15<2:29:58,  1.82s/it]
 66%|██████▋   | 9785/14725 [4:59:17<2:28:39,  1.81s/it]
 66%|██████▋   | 9786/14725 [4:59:18<2:30:13,  1.82s/it]
 66%|██████▋   | 9787/14725 [4:59:20<2:29:03,  1.81s/it]
 66%|██████▋   
+0: {'loss': 0.7926, 'grad_norm': 0.704989289575352, 'learning_rate': 2e-05, 'epoch': 0.66}
+0: | 9788/14725 [4:59:22<2:28:42,  1.81s/it]
 66%|██████▋   | 9789/14725 [4:59:24<2:33:09,  1.86s/it]
 66%|██████▋   | 9790/14725 [4:59:26<2:32:21,  1.85s/it]
                                                        

 66%|██████▋   | 9790/14725 [4:59:26<2:32:21,  1.85s/it]
 66%|██████▋   | 9791/14725 [4:59:28<2:31:54,  1.85s/it]
 66%|██████▋   | 9792/14725 [4:59:29<2:30:09,  1.83s/it]
 67%|██████▋   | 9793/14725 [4:59:31<2:30:15,  1.83s/it]
 67%|██████▋   | 9794/14725 [4:59:33<2:28:29,  1.81s/it]
 67%|██████▋   | 9795/14725 [4:59:35<2:27:04,  1.79s/it]
 67%|██████▋   | 9796/14725 [4:59:37<2:27:44,  1.80s/it]
 67%|██████▋   | 9797/14725 [4:59:38<2:27:51,  1.80s/it]
 67%|██████▋   | 9798/14725 [4:59:40<2:27:20,  1.79s/it]
 67%|██████▋   | 9799/14725 [4:59:42<2:26:03,  1.78s/it]
 67%|██████▋   | 9800/14725 [4:59:44<2:25:51,  1.78s/it]
 
+0: {'loss': 0.8073, 'grad_norm': 0.6326078406076353, 'learning_rate': 2e-05, 'epoch': 0.67}
+0: {'loss': 0.7986, 'grad_norm': 0.7173050022118481, 'learning_rate': 2e-05, 'epoch': 0.67}
+0:                                                        

 67%|██████▋   | 9800/14725 [4:59:44<2:25:51,  1.78s/it]
 67%|██████▋   | 9801/14725 [4:59:46<2:27:32,  1.80s/it]
 67%|██████▋   | 9802/14725 [4:59:48<2:30:46,  1.84s/it]
 67%|██████▋   | 9803/14725 [4:59:49<2:28:46,  1.81s/it]
 67%|██████▋   | 9804/14725 [4:59:51<2:28:32,  1.81s/it]
 67%|██████▋   | 9805/14725 [4:59:53<2:29:17,  1.82s/it]
 67%|██████▋   | 9806/14725 [4:59:55<2:27:41,  1.80s/it]
 67%|██████▋   | 9807/14725 [4:59:56<2:27:37,  1.80s/it]
 67%|██████▋   | 9808/14725 [4:59:58<2:28:40,  1.81s/it]
 67%|██████▋   | 9809/14725 [5:00:00<2:29:19,  1.82s/it]
 67%|██████▋   | 9810/14725 [5:00:02<2:28:31,  1.81s/it]
                                                        

 67%|██████▋   | 9810/14725 [5:00:02<2:28:31,  1.81s/it]
 67%|██████▋   | 9811/14725 [5:00:04<2:28:1
+0: {'loss': 0.779, 'grad_norm': 0.6875905928786444, 'learning_rate': 2e-05, 'epoch': 0.67}
+0: 4,  1.81s/it]
 67%|██████▋   | 9812/14725 [5:00:06<2:29:08,  1.82s/it]
 67%|██████▋   | 9813/14725 [5:00:07<2:27:47,  1.81s/it]
 67%|██████▋   | 9814/14725 [5:00:09<2:26:49,  1.79s/it]
 67%|██████▋   | 9815/14725 [5:00:11<2:26:19,  1.79s/it]
 67%|██████▋   | 9816/14725 [5:00:13<2:25:15,  1.78s/it]
 67%|██████▋   | 9817/14725 [5:00:14<2:25:27,  1.78s/it]
 67%|██████▋   | 9818/14725 [5:00:16<2:25:16,  1.78s/it]
 67%|██████▋   | 9819/14725 [5:00:18<2:25:26,  1.78s/it]
 67%|██████▋   | 9820/14725 [5:00:20<2:25:29,  1.78s/it]
                                                        

 67%|██████▋   | 9820/14725 [5:00:20<2:25:29,  1.78s/it]
 67%|██████▋   | 9821/14725 [5:00:22<2:25:56,  1.79s/it]
 67%|██████▋   | 9822/14725 [5:00:23<2:25:20,  1.78s/it]
 67%|██████▋   | 9823/14725 [5:00:25<2:26:02,  1.79s/it]
 67%|██████▋   
+0: {'loss': 0.8194, 'grad_norm': 0.6573815134334309, 'learning_rate': 2e-05, 'epoch': 0.67}
+0: | 9824/14725 [5:00:27<2:25:42,  1.78s/it]
 67%|██████▋   | 9825/14725 [5:00:29<2:26:24,  1.79s/it]
 67%|██████▋   | 9826/14725 [5:00:31<2:25:39,  1.78s/it]
 67%|██████▋   | 9827/14725 [5:00:32<2:24:33,  1.77s/it]
 67%|██████▋   | 9828/14725 [5:00:34<2:26:13,  1.79s/it]
 67%|██████▋   | 9829/14725 [5:00:36<2:27:58,  1.81s/it]
 67%|██████▋   | 9830/14725 [5:00:38<2:26:23,  1.79s/it]
                                                        

 67%|██████▋   | 9830/14725 [5:00:38<2:26:23,  1.79s/it]
 67%|██████▋   | 9831/14725 [5:00:40<2:27:33,  1.81s/it]
 67%|██████▋   | 9832/14725 [5:00:41<2:27:54,  1.81s/it]
 67%|██████▋   | 9833/14725 [5:00:43<2:27:35,  1.81s/it]
 67%|██████▋   | 9834/14725 [5:00:45<2:26:14,  1.79s/it]
 67%|██████▋   | 9835/14725 [5:00:47<2:25:55,  1.79s/it]
 67%|██████▋   | 9836/14725 [5:00:48<2:25:16,  1.78s/it]
 
+0: {'loss': 0.803, 'grad_norm': 0.8185724269666471, 'learning_rate': 2e-05, 'epoch': 0.67}
+0: 67%|██████▋   | 9837/14725 [5:00:50<2:25:44,  1.79s/it]
 67%|██████▋   | 9838/14725 [5:00:52<2:26:21,  1.80s/it]
 67%|██████▋   | 9839/14725 [5:00:54<2:26:13,  1.80s/it]
 67%|██████▋   | 9840/14725 [5:00:56<2:26:43,  1.80s/it]
                                                        

 67%|██████▋   | 9840/14725 [5:00:56<2:26:43,  1.80s/it]
 67%|██████▋   | 9841/14725 [5:00:58<2:26:50,  1.80s/it]
 67%|██████▋   | 9842/14725 [5:00:59<2:26:32,  1.80s/it]
 67%|██████▋   | 9843/14725 [5:01:01<2:26:18,  1.80s/it]
 67%|██████▋   | 9844/14725 [5:01:03<2:25:02,  1.78s/it]
 67%|██████▋   | 9845/14725 [5:01:05<2:25:44,  1.79s/it]
 67%|██████▋   | 9846/14725 [5:01:07<2:29:07,  1.83s/it]
 67%|██████▋   | 9847/14725 [5:01:08<2:28:59,  1.83s/it]
 67%|██████▋   | 9848/14725 [5:01:10<2:26:54,  1.81s/it]
 67%|██████▋   | 9849/14725 [5
+0: {'loss': 0.8188, 'grad_norm': 0.7311493533638829, 'learning_rate': 2e-05, 'epoch': 0.67}
+0: {'loss': 0.7868, 'grad_norm': 0.667867646902766, 'learning_rate': 2e-05, 'epoch': 0.67}
+0: :01:12<2:28:48,  1.83s/it]
 67%|██████▋   | 9850/14725 [5:01:14<2:26:28,  1.80s/it]
                                                        

 67%|██████▋   | 9850/14725 [5:01:14<2:26:28,  1.80s/it]
 67%|██████▋   | 9851/14725 [5:01:16<2:26:44,  1.81s/it]
 67%|██████▋   | 9852/14725 [5:01:17<2:28:05,  1.82s/it]
 67%|██████▋   | 9853/14725 [5:01:19<2:26:52,  1.81s/it]
 67%|██████▋   | 9854/14725 [5:01:21<2:25:34,  1.79s/it]
 67%|██████▋   | 9855/14725 [5:01:23<2:26:41,  1.81s/it]
 67%|██████▋   | 9856/14725 [5:01:25<2:25:12,  1.79s/it]
 67%|██████▋   | 9857/14725 [5:01:26<2:24:50,  1.79s/it]
 67%|██████▋   | 9858/14725 [5:01:28<2:26:03,  1.80s/it]
 67%|██████▋   | 9859/14725 [5:01:30<2:24:59,  1.79s/it]
 67%|██████▋   | 9860/14725 [5:01:32<2:27:07,  1.81s/it]
                                                        

 67%|██████▋   
+0: {'loss': 0.7889, 'grad_norm': 0.7389787571292566, 'learning_rate': 2e-05, 'epoch': 0.67}
+0: | 9860/14725 [5:01:32<2:27:07,  1.81s/it]
 67%|██████▋   | 9861/14725 [5:01:34<2:27:22,  1.82s/it]
 67%|██████▋   | 9862/14725 [5:01:35<2:25:43,  1.80s/it]
 67%|██████▋   | 9863/14725 [5:01:37<2:27:47,  1.82s/it]
 67%|██████▋   | 9864/14725 [5:01:39<2:27:26,  1.82s/it]
 67%|██████▋   | 9865/14725 [5:01:41<2:28:00,  1.83s/it]
 67%|██████▋   | 9866/14725 [5:01:43<2:28:08,  1.83s/it]
 67%|██████▋   | 9867/14725 [5:01:45<2:28:16,  1.83s/it]
 67%|██████▋   | 9868/14725 [5:01:46<2:27:35,  1.82s/it]
 67%|██████▋   | 9869/14725 [5:01:48<2:25:50,  1.80s/it]
 67%|██████▋   | 9870/14725 [5:01:50<2:25:26,  1.80s/it]
                                                        

 67%|██████▋   | 9870/14725 [5:01:50<2:25:26,  1.80s/it]
 67%|██████▋   | 9871/14725 [5:01:52<2:24:45,  1.79s/it]
 67%|██████▋   | 9872/14725 [5:01:54<2:25:42,  1.80s/it]
 
+0: {'loss': 0.8038, 'grad_norm': 0.7655446952301249, 'learning_rate': 2e-05, 'epoch': 0.67}
+0: 67%|██████▋   | 9873/14725 [5:01:55<2:26:35,  1.81s/it]
 67%|██████▋   | 9874/14725 [5:01:57<2:26:56,  1.82s/it]
 67%|██████▋   | 9875/14725 [5:01:59<2:25:27,  1.80s/it]
 67%|██████▋   | 9876/14725 [5:02:01<2:25:19,  1.80s/it]
 67%|██████▋   | 9877/14725 [5:02:03<2:24:09,  1.78s/it]
 67%|██████▋   | 9878/14725 [5:02:04<2:24:04,  1.78s/it]
 67%|██████▋   | 9879/14725 [5:02:06<2:22:56,  1.77s/it]
 67%|██████▋   | 9880/14725 [5:02:08<2:23:55,  1.78s/it]
                                                        

 67%|██████▋   | 9880/14725 [5:02:08<2:23:55,  1.78s/it]
 67%|██████▋   | 9881/14725 [5:02:10<2:25:26,  1.80s/it]
 67%|██████▋   | 9882/14725 [5:02:12<2:25:42,  1.81s/it]
 67%|██████▋   | 9883/14725 [5:02:13<2:26:41,  1.82s/it]
 67%|██████▋   | 9884/14725 [5:02:15<2:26:02,  1.81s/it]
 67%|██████▋   | 9885/14725 [5
+0: {'loss': 0.8065, 'grad_norm': 0.7316530324331317, 'learning_rate': 2e-05, 'epoch': 0.67}
+0: :02:17<2:27:23,  1.83s/it]
 67%|██████▋   | 9886/14725 [5:02:19<2:26:51,  1.82s/it]
 67%|██████▋   | 9887/14725 [5:02:21<2:25:13,  1.80s/it]
 67%|██████▋   | 9888/14725 [5:02:22<2:25:14,  1.80s/it]
 67%|██████▋   | 9889/14725 [5:02:24<2:26:05,  1.81s/it]
 67%|██████▋   | 9890/14725 [5:02:26<2:26:09,  1.81s/it]
                                                        

 67%|██████▋   | 9890/14725 [5:02:26<2:26:09,  1.81s/it]
 67%|██████▋   | 9891/14725 [5:02:28<2:25:13,  1.80s/it]
 67%|██████▋   | 9892/14725 [5:02:30<2:25:26,  1.81s/it]
 67%|██████▋   | 9893/14725 [5:02:31<2:26:28,  1.82s/it]
 67%|██████▋   | 9894/14725 [5:02:34<2:31:57,  1.89s/it]
 67%|██████▋   | 9895/14725 [5:02:35<2:31:41,  1.88s/it]
 67%|██████▋   | 9896/14725 [5:02:37<2:30:15,  1.87s/it]
 67%|██████▋   | 9897/14725 [5:02:39<2:29:35,  1.86s/it]
 67%|███�
+0: {'loss': 0.8092, 'grad_norm': 0.7214076065780094, 'learning_rate': 2e-05, 'epoch': 0.67}
+0: �██▋   | 9898/14725 [5:02:41<2:28:40,  1.85s/it]
 67%|██████▋   | 9899/14725 [5:02:43<2:28:20,  1.84s/it]
 67%|██████▋   | 9900/14725 [5:02:45<2:26:38,  1.82s/it]
                                                        

 67%|██████▋   | 9900/14725 [5:02:45<2:26:38,  1.82s/it]
 67%|██████▋   | 9901/14725 [5:02:46<2:27:01,  1.83s/it]
 67%|██████▋   | 9902/14725 [5:02:48<2:27:06,  1.83s/it]
 67%|██████▋   | 9903/14725 [5:02:50<2:26:11,  1.82s/it]
 67%|██████▋   | 9904/14725 [5:02:52<2:26:03,  1.82s/it]
 67%|██████▋   | 9905/14725 [5:02:54<2:25:42,  1.81s/it]
 67%|██████▋   | 9906/14725 [5:02:55<2:24:13,  1.80s/it]
 67%|██████▋   | 9907/14725 [5:02:57<2:26:13,  1.82s/it]
 67%|██████▋   | 9908/14725 [5:02:59<2:28:24,  1.85s/it]
 67%|██████▋   | 9909/14725 [5:03:01<2:27:16,  1.83s/it]
 67%|██████▋   | 9910/14725 [5:03:03<2:26:46,
+0: {'loss': 0.8023, 'grad_norm': 0.6879697343719868, 'learning_rate': 2e-05, 'epoch': 0.67}
+0: {'loss': 0.7922, 'grad_norm': 0.7112169310257284, 'learning_rate': 2e-05, 'epoch': 0.67}
+0:   1.83s/it]
                                                        

 67%|██████▋   | 9910/14725 [5:03:03<2:26:46,  1.83s/it]
 67%|██████▋   | 9911/14725 [5:03:05<2:26:03,  1.82s/it]
 67%|██████▋   | 9912/14725 [5:03:06<2:25:58,  1.82s/it]
 67%|██████▋   | 9913/14725 [5:03:08<2:25:23,  1.81s/it]
 67%|██████▋   | 9914/14725 [5:03:10<2:25:14,  1.81s/it]
 67%|██████▋   | 9915/14725 [5:03:12<2:25:51,  1.82s/it]
 67%|██████▋   | 9916/14725 [5:03:14<2:24:25,  1.80s/it]
 67%|██████▋   | 9917/14725 [5:03:15<2:26:18,  1.83s/it]
 67%|██████▋   | 9918/14725 [5:03:17<2:25:14,  1.81s/it]
 67%|██████▋   | 9919/14725 [5:03:19<2:23:38,  1.79s/it]
 67%|██████▋   | 9920/14725 [5:03:21<2:24:39,  1.81s/it]
                                                        

 67%|██████▋   | 9920/14725 [5:03:21<2:24:39,  1.81s/it]
 67%|██████▋   | 9921/14725 [5
+0: {'loss': 0.7989, 'grad_norm': 0.6546262155308483, 'learning_rate': 2e-05, 'epoch': 0.67}
+0: :03:23<2:24:37,  1.81s/it]
 67%|██████▋   | 9922/14725 [5:03:24<2:24:19,  1.80s/it]
 67%|██████▋   | 9923/14725 [5:03:26<2:23:18,  1.79s/it]
 67%|██████▋   | 9924/14725 [5:03:28<2:24:44,  1.81s/it]
 67%|██████▋   | 9925/14725 [5:03:30<2:23:08,  1.79s/it]
 67%|██████▋   | 9926/14725 [5:03:32<2:22:30,  1.78s/it]
 67%|██████▋   | 9927/14725 [5:03:33<2:24:31,  1.81s/it]
 67%|██████▋   | 9928/14725 [5:03:35<2:22:55,  1.79s/it]
 67%|██████▋   | 9929/14725 [5:03:37<2:26:59,  1.84s/it]
 67%|██████▋   | 9930/14725 [5:03:39<2:26:04,  1.83s/it]
                                                        

 67%|██████▋   | 9930/14725 [5:03:39<2:26:04,  1.83s/it]
 67%|██████▋   | 9931/14725 [5:03:41<2:25:17,  1.82s/it]
 67%|██████▋   | 9932/14725 [5:03:42<2:23:52,  1.80s/it]
 67%|██████▋   | 9933/14725 [5:03:44<2:23:46,  1.80s/it]
 67%|███�
+0: {'loss': 0.7989, 'grad_norm': 0.682699831025369, 'learning_rate': 2e-05, 'epoch': 0.68}
+0: �██▋   | 9934/14725 [5:03:46<2:23:47,  1.80s/it]
 67%|██████▋   | 9935/14725 [5:03:48<2:25:56,  1.83s/it]
 67%|██████▋   | 9936/14725 [5:03:50<2:24:17,  1.81s/it]
 67%|██████▋   | 9937/14725 [5:03:52<2:25:53,  1.83s/it]
 67%|██████▋   | 9938/14725 [5:03:54<2:28:49,  1.87s/it]
 67%|██████▋   | 9939/14725 [5:03:55<2:27:45,  1.85s/it]
 68%|██████▊   | 9940/14725 [5:03:57<2:27:19,  1.85s/it]
                                                        

 68%|██████▊   | 9940/14725 [5:03:57<2:27:19,  1.85s/it]
 68%|██████▊   | 9941/14725 [5:03:59<2:27:18,  1.85s/it]
 68%|██████▊   | 9942/14725 [5:04:01<2:25:52,  1.83s/it]
 68%|██████▊   | 9943/14725 [5:04:03<2:32:14,  1.91s/it]
 68%|██████▊   | 9944/14725 [5:04:05<2:30:06,  1.88s/it]
 68%|██████▊   | 9945/14725 [5:04:07<2:28:36,  1.87s/it]
 68%|██████▊   | 9946/14725 [5:04:08<2:28:41,
+0: {'loss': 0.8341, 'grad_norm': 0.7368174003026523, 'learning_rate': 2e-05, 'epoch': 0.68}
+0:   1.87s/it]
 68%|██████▊   | 9947/14725 [5:04:10<2:27:07,  1.85s/it]
 68%|██████▊   | 9948/14725 [5:04:12<2:25:49,  1.83s/it]
 68%|██████▊   | 9949/14725 [5:04:14<2:25:50,  1.83s/it]
 68%|██████▊   | 9950/14725 [5:04:16<2:24:14,  1.81s/it]
                                                        

 68%|██████▊   | 9950/14725 [5:04:16<2:24:14,  1.81s/it]
 68%|██████▊   | 9951/14725 [5:04:18<2:28:50,  1.87s/it]
 68%|██████▊   | 9952/14725 [5:04:20<2:29:29,  1.88s/it]
 68%|██████▊   | 9953/14725 [5:04:21<2:28:47,  1.87s/it]
 68%|██████▊   | 9954/14725 [5:04:23<2:28:46,  1.87s/it]
 68%|██████▊   | 9955/14725 [5:04:25<2:29:01,  1.87s/it]
 68%|██████▊   | 9956/14725 [5:04:27<2:26:41,  1.85s/it]
 68%|██████▊   | 9957/14725 [5:04:29<2:24:43,  1.82s/it]
 68%|██████▊   | 9958/14725 [5:04:30<2:23:27,  1.81s/it]
 68%|██████▊   | 
+0: {'loss': 0.7884, 'grad_norm': 0.7515105803687553, 'learning_rate': 2e-05, 'epoch': 0.68}
+0: {'loss': 0.8156, 'grad_norm': 0.7231020233365056, 'learning_rate': 2e-05, 'epoch': 0.68}
+0: 9959/14725 [5:04:32<2:23:48,  1.81s/it]
 68%|██████▊   | 9960/14725 [5:04:34<2:28:16,  1.87s/it]
                                                        

 68%|██████▊   | 9960/14725 [5:04:34<2:28:16,  1.87s/it]
 68%|██████▊   | 9961/14725 [5:04:36<2:27:00,  1.85s/it]
 68%|██████▊   | 9962/14725 [5:04:38<2:24:44,  1.82s/it]
 68%|██████▊   | 9963/14725 [5:04:40<2:24:19,  1.82s/it]
 68%|██████▊   | 9964/14725 [5:04:42<2:27:18,  1.86s/it]
 68%|██████▊   | 9965/14725 [5:04:43<2:27:07,  1.85s/it]
 68%|██████▊   | 9966/14725 [5:04:45<2:26:22,  1.85s/it]
 68%|██████▊   | 9967/14725 [5:04:47<2:25:47,  1.84s/it]
 68%|██████▊   | 9968/14725 [5:04:49<2:25:49,  1.84s/it]
 68%|██████▊   | 9969/14725 [5:04:51<2:24:18,  1.82s/it]
 68%|██████▊   | 9970/14725 [5:04:53<2:23:13,  1.81s/it]
                                                        

 68%|███�
+0: {'loss': 0.7997, 'grad_norm': 0.6975172261470389, 'learning_rate': 2e-05, 'epoch': 0.68}
+0: �██▊   | 9970/14725 [5:04:53<2:23:13,  1.81s/it]
 68%|██████▊   | 9971/14725 [5:04:54<2:22:09,  1.79s/it]
 68%|██████▊   | 9972/14725 [5:04:56<2:21:13,  1.78s/it]
 68%|██████▊   | 9973/14725 [5:04:58<2:22:35,  1.80s/it]
 68%|██████▊   | 9974/14725 [5:05:00<2:21:26,  1.79s/it]
 68%|██████▊   | 9975/14725 [5:05:01<2:20:21,  1.77s/it]
 68%|██████▊   | 9976/14725 [5:05:03<2:22:42,  1.80s/it]
 68%|██████▊   | 9977/14725 [5:05:05<2:22:05,  1.80s/it]
 68%|██████▊   | 9978/14725 [5:05:07<2:21:16,  1.79s/it]
 68%|██████▊   | 9979/14725 [5:05:09<2:20:35,  1.78s/it]
 68%|██████▊   | 9980/14725 [5:05:10<2:21:41,  1.79s/it]
                                                        

 68%|██████▊   | 9980/14725 [5:05:10<2:21:41,  1.79s/it]
 68%|██████▊   | 9981/14725 [5:05:12<2:21:33,  1.79s/it]
 68%|██████▊   | 9982/14725 [5:05:14<2:24:03,
+0: {'loss': 0.7954, 'grad_norm': 0.7073868437475451, 'learning_rate': 2e-05, 'epoch': 0.68}
+0:   1.82s/it]
 68%|██████▊   | 9983/14725 [5:05:16<2:29:11,  1.89s/it]
 68%|██████▊   | 9984/14725 [5:05:18<2:26:32,  1.85s/it]
 68%|██████▊   | 9985/14725 [5:05:20<2:26:24,  1.85s/it]
 68%|██████▊   | 9986/14725 [5:05:22<2:24:56,  1.84s/it]
 68%|██████▊   | 9987/14725 [5:05:23<2:23:27,  1.82s/it]
 68%|██████▊   | 9988/14725 [5:05:25<2:22:59,  1.81s/it]
 68%|██████▊   | 9989/14725 [5:05:27<2:24:01,  1.82s/it]
 68%|██████▊   | 9990/14725 [5:05:29<2:24:24,  1.83s/it]
                                                        

 68%|██████▊   | 9990/14725 [5:05:29<2:24:24,  1.83s/it]
 68%|██████▊   | 9991/14725 [5:05:31<2:23:01,  1.81s/it]
 68%|██████▊   | 9992/14725 [5:05:32<2:21:57,  1.80s/it]
 68%|██████▊   | 9993/14725 [5:05:34<2:21:04,  1.79s/it]
 68%|██████▊   | 9994/14725 [5:05:36<2:21:17,  1.79s/it]
 68%|██████▊   | 
+0: {'loss': 0.7869, 'grad_norm': 0.6713308695279653, 'learning_rate': 2e-05, 'epoch': 0.68}
+0: 9995/14725 [5:05:38<2:20:57,  1.79s/it]
 68%|██████▊   | 9996/14725 [5:05:39<2:19:44,  1.77s/it]
 68%|██████▊   | 9997/14725 [5:05:41<2:20:15,  1.78s/it]
 68%|██████▊   | 9998/14725 [5:05:43<2:20:08,  1.78s/it]
 68%|██████▊   | 9999/14725 [5:05:45<2:20:46,  1.79s/it]
 68%|██████▊   | 10000/14725 [5:05:47<2:20:28,  1.78s/it]
                                                         

 68%|██████▊   | 10000/14725 [5:05:47<2:20:28,  1.78s/it]
 68%|██████▊   | 10001/14725 [5:05:48<2:19:57,  1.78s/it]
 68%|██████▊   | 10002/14725 [5:05:50<2:19:18,  1.77s/it]
 68%|██████▊   | 10003/14725 [5:05:52<2:19:25,  1.77s/it]
 68%|██████▊   | 10004/14725 [5:05:54<2:19:37,  1.77s/it]
 68%|██████▊   | 10005/14725 [5:05:55<2:20:12,  1.78s/it]
 68%|██████▊   | 10006/14725 [5:05:57<2:21:41,  1.80s/it]
 68%|██████▊   | 10007/14725 [5:05:59<2:21:14,  1.8
+0: {'loss': 0.8113, 'grad_norm': 0.7254150129794148, 'learning_rate': 2e-05, 'epoch': 0.68}
+0: 0s/it]
 68%|██████▊   | 10008/14725 [5:06:01<2:21:02,  1.79s/it]
 68%|██████▊   | 10009/14725 [5:06:03<2:20:33,  1.79s/it]
 68%|██████▊   | 10010/14725 [5:06:04<2:20:57,  1.79s/it]
                                                         

 68%|██████▊   | 10010/14725 [5:06:04<2:20:57,  1.79s/it]
 68%|██████▊   | 10011/14725 [5:06:06<2:21:49,  1.81s/it]
 68%|██████▊   | 10012/14725 [5:06:08<2:21:25,  1.80s/it]
 68%|██████▊   | 10013/14725 [5:06:10<2:21:44,  1.80s/it]
 68%|██████▊   | 10014/14725 [5:06:12<2:21:17,  1.80s/it]
 68%|██████▊   | 10015/14725 [5:06:13<2:21:17,  1.80s/it]
 68%|██████▊   | 10016/14725 [5:06:15<2:20:49,  1.79s/it]
 68%|██████▊   | 10017/14725 [5:06:17<2:19:49,  1.78s/it]
 68%|██████▊   | 10018/14725 [5:06:19<2:22:02,  1.81s/it]
 68%|██████▊   | 10019/14725 [5:06:21<2:21:12,  1.80s/it]
 68%|█████�
+0: {'loss': 0.7875, 'grad_norm': 0.713828598446685, 'learning_rate': 2e-05, 'epoch': 0.68}
+0: {'loss': 0.8124, 'grad_norm': 0.6790198922205415, 'learning_rate': 2e-05, 'epoch': 0.68}
+0: �▊   | 10020/14725 [5:06:23<2:23:05,  1.82s/it]
                                                         

 68%|██████▊   | 10020/14725 [5:06:23<2:23:05,  1.82s/it]
 68%|██████▊   | 10021/14725 [5:06:24<2:22:53,  1.82s/it]
 68%|██████▊   | 10022/14725 [5:06:26<2:25:02,  1.85s/it]
 68%|██████▊   | 10023/14725 [5:06:28<2:23:13,  1.83s/it]
 68%|██████▊   | 10024/14725 [5:06:30<2:22:19,  1.82s/it]
 68%|██████▊   | 10025/14725 [5:06:32<2:20:58,  1.80s/it]
 68%|██████▊   | 10026/14725 [5:06:33<2:19:35,  1.78s/it]
 68%|██████▊   | 10027/14725 [5:06:35<2:20:28,  1.79s/it]
 68%|██████▊   | 10028/14725 [5:06:37<2:20:12,  1.79s/it]
 68%|██████▊   | 10029/14725 [5:06:39<2:19:29,  1.78s/it]
 68%|██████▊   | 10030/14725 [5:06:40<2:19:35,  1.78s/it]
                                                         

 68%|██████▊   | 10030/14725 [5:06:41<2:19:35,  1.
+0: {'loss': 0.8224, 'grad_norm': 0.6670032619877255, 'learning_rate': 2e-05, 'epoch': 0.68}
+0: 78s/it]
 68%|██████▊   | 10031/14725 [5:06:42<2:19:11,  1.78s/it]
 68%|██████▊   | 10032/14725 [5:06:44<2:20:52,  1.80s/it]
 68%|██████▊   | 10033/14725 [5:06:46<2:21:02,  1.80s/it]
 68%|██████▊   | 10034/14725 [5:06:48<2:21:47,  1.81s/it]
 68%|██████▊   | 10035/14725 [5:06:50<2:21:37,  1.81s/it]
 68%|██████▊   | 10036/14725 [5:06:51<2:22:08,  1.82s/it]
 68%|██████▊   | 10037/14725 [5:06:53<2:22:35,  1.82s/it]
 68%|██████▊   | 10038/14725 [5:06:55<2:21:52,  1.82s/it]
 68%|██████▊   | 10039/14725 [5:06:57<2:20:47,  1.80s/it]
 68%|██████▊   | 10040/14725 [5:06:59<2:21:44,  1.82s/it]
                                                         

 68%|██████▊   | 10040/14725 [5:06:59<2:21:44,  1.82s/it]
 68%|██████▊   | 10041/14725 [5:07:01<2:22:41,  1.83s/it]
 68%|██████▊   | 10042/14725 [5:07:02<2:23:22,  1.84s/it]
 68%|█████�
+0: {'loss': 0.8149, 'grad_norm': 0.70545566584892, 'learning_rate': 2e-05, 'epoch': 0.68}
+0: ��▊   | 10043/14725 [5:07:04<2:24:41,  1.85s/it]
 68%|██████▊   | 10044/14725 [5:07:06<2:22:58,  1.83s/it]
 68%|██████▊   | 10045/14725 [5:07:08<2:24:16,  1.85s/it]
 68%|██████▊   | 10046/14725 [5:07:10<2:22:12,  1.82s/it]
 68%|██████▊   | 10047/14725 [5:07:12<2:22:46,  1.83s/it]
 68%|██████▊   | 10048/14725 [5:07:13<2:21:04,  1.81s/it]
 68%|██████▊   | 10049/14725 [5:07:15<2:20:42,  1.81s/it]
 68%|██████▊   | 10050/14725 [5:07:17<2:19:27,  1.79s/it]
                                                         

 68%|██████▊   | 10050/14725 [5:07:17<2:19:27,  1.79s/it]
 68%|██████▊   | 10051/14725 [5:07:19<2:19:17,  1.79s/it]
 68%|██████▊   | 10052/14725 [5:07:20<2:18:37,  1.78s/it]
 68%|██████▊   | 10053/14725 [5:07:22<2:18:03,  1.77s/it]
 68%|██████▊   | 10054/14725 [5:07:24<2:18:37,  1.78s/it]
 68%|██████▊   | 10055/14725 [5:07:2
+0: {'loss': 0.8171, 'grad_norm': 0.6741960781920029, 'learning_rate': 2e-05, 'epoch': 0.68}
+0: 6<2:19:06,  1.79s/it]
 68%|██████▊   | 10056/14725 [5:07:28<2:19:56,  1.80s/it]
 68%|██████▊   | 10057/14725 [5:07:29<2:20:02,  1.80s/it]
 68%|██████▊   | 10058/14725 [5:07:31<2:21:11,  1.82s/it]
 68%|██████▊   | 10059/14725 [5:07:33<2:21:49,  1.82s/it]
 68%|██████▊   | 10060/14725 [5:07:35<2:21:50,  1.82s/it]
                                                         

 68%|██████▊   | 10060/14725 [5:07:35<2:21:50,  1.82s/it]
 68%|██████▊   | 10061/14725 [5:07:37<2:21:49,  1.82s/it]
 68%|██████▊   | 10062/14725 [5:07:39<2:21:34,  1.82s/it]
 68%|██████▊   | 10063/14725 [5:07:40<2:20:42,  1.81s/it]
 68%|██████▊   | 10064/14725 [5:07:42<2:21:19,  1.82s/it]
 68%|██████▊   | 10065/14725 [5:07:44<2:20:10,  1.80s/it]
 68%|██████▊   | 10066/14725 [5:07:46<2:21:04,  1.82s/it]
 68%|██████▊   | 10067/14725 [5:07:48<2:21:48,  1.83s/it]
 68%|�
+0: {'loss': 0.8291, 'grad_norm': 0.7070156367325162, 'learning_rate': 2e-05, 'epoch': 0.68}
+0: �█████▊   | 10068/14725 [5:07:50<2:23:07,  1.84s/it]
 68%|██████▊   | 10069/14725 [5:07:51<2:24:23,  1.86s/it]
 68%|██████▊   | 10070/14725 [5:07:53<2:22:05,  1.83s/it]
                                                         

 68%|██████▊   | 10070/14725 [5:07:53<2:22:05,  1.83s/it]
 68%|██████▊   | 10071/14725 [5:07:55<2:21:22,  1.82s/it]
 68%|██████▊   | 10072/14725 [5:07:57<2:22:24,  1.84s/it]
 68%|██████▊   | 10073/14725 [5:07:59<2:20:47,  1.82s/it]
 68%|██████▊   | 10074/14725 [5:08:00<2:18:57,  1.79s/it]
 68%|██████▊   | 10075/14725 [5:08:02<2:17:50,  1.78s/it]
 68%|██████▊   | 10076/14725 [5:08:04<2:17:34,  1.78s/it]
 68%|██████▊   | 10077/14725 [5:08:06<2:17:41,  1.78s/it]
 68%|██████▊   | 10078/14725 [5:08:07<2:17:39,  1.78s/it]
 68%|██████▊   | 10079/14725 [5:08:09<2:17:45,  1.78s/it]
 68%|██████▊   | 10080
+0: {'loss': 0.8042, 'grad_norm': 0.6762128508374604, 'learning_rate': 2e-05, 'epoch': 0.68}
+0: {'loss': 0.8265, 'grad_norm': 0.6785083621091185, 'learning_rate': 2e-05, 'epoch': 0.69}
+0: /14725 [5:08:11<2:17:11,  1.77s/it]
                                                         

 68%|██████▊   | 10080/14725 [5:08:11<2:17:11,  1.77s/it]
 68%|██████▊   | 10081/14725 [5:08:13<2:17:14,  1.77s/it]
 68%|██████▊   | 10082/14725 [5:08:15<2:21:15,  1.83s/it]
 68%|██████▊   | 10083/14725 [5:08:17<2:20:49,  1.82s/it]
 68%|██████▊   | 10084/14725 [5:08:18<2:19:37,  1.81s/it]
 68%|██████▊   | 10085/14725 [5:08:20<2:19:45,  1.81s/it]
 68%|██████▊   | 10086/14725 [5:08:22<2:18:56,  1.80s/it]
 69%|██████▊   | 10087/14725 [5:08:24<2:18:44,  1.79s/it]
 69%|██████▊   | 10088/14725 [5:08:25<2:18:10,  1.79s/it]
 69%|██████▊   | 10089/14725 [5:08:27<2:19:00,  1.80s/it]
 69%|██████▊   | 10090/14725 [5:08:29<2:19:23,  1.80s/it]
                                                         

 69%|██████▊   | 10090/14725 [5:08:29<2:19:23,  1.80s/it]
 69%|�
+0: {'loss': 0.795, 'grad_norm': 0.6581293525022646, 'learning_rate': 2e-05, 'epoch': 0.69}
+0: ��█████▊   | 10091/14725 [5:08:31<2:19:18,  1.80s/it]
 69%|██████▊   | 10092/14725 [5:08:33<2:18:18,  1.79s/it]
 69%|██████▊   | 10093/14725 [5:08:34<2:18:27,  1.79s/it]
 69%|██████▊   | 10094/14725 [5:08:36<2:17:14,  1.78s/it]
 69%|██████▊   | 10095/14725 [5:08:38<2:18:02,  1.79s/it]
 69%|██████▊   | 10096/14725 [5:08:40<2:17:41,  1.78s/it]
 69%|██████▊   | 10097/14725 [5:08:42<2:16:51,  1.77s/it]
 69%|██████▊   | 10098/14725 [5:08:43<2:16:37,  1.77s/it]
 69%|██████▊   | 10099/14725 [5:08:45<2:17:26,  1.78s/it]
 69%|██████▊   | 10100/14725 [5:08:47<2:18:07,  1.79s/it]
                                                         

 69%|██████▊   | 10100/14725 [5:08:47<2:18:07,  1.79s/it]
 69%|██████▊   | 10101/14725 [5:08:49<2:19:48,  1.81s/it]
 69%|██████▊   | 10102/14725 [5:08:51<2:19:01,  1.80s/it]
 69%|██████▊   | 1010
+0: {'loss': 0.7918, 'grad_norm': 0.6608702641721893, 'learning_rate': 2e-05, 'epoch': 0.69}
+0: 3/14725 [5:08:52<2:17:59,  1.79s/it]
 69%|██████▊   | 10104/14725 [5:08:54<2:18:25,  1.80s/it]
 69%|██████▊   | 10105/14725 [5:08:56<2:18:40,  1.80s/it]
 69%|██████▊   | 10106/14725 [5:08:58<2:19:16,  1.81s/it]
 69%|██████▊   | 10107/14725 [5:09:00<2:18:11,  1.80s/it]
 69%|██████▊   | 10108/14725 [5:09:01<2:18:23,  1.80s/it]
 69%|██████▊   | 10109/14725 [5:09:03<2:18:26,  1.80s/it]
 69%|██████▊   | 10110/14725 [5:09:05<2:17:32,  1.79s/it]
                                                         

 69%|██████▊   | 10110/14725 [5:09:05<2:17:32,  1.79s/it]
 69%|██████▊   | 10111/14725 [5:09:07<2:17:38,  1.79s/it]
 69%|██████▊   | 10112/14725 [5:09:09<2:18:33,  1.80s/it]
 69%|██████▊   | 10113/14725 [5:09:10<2:18:40,  1.80s/it]
 69%|██████▊   | 10114/14725 [5:09:12<2:18:52,  1.81s/it]
 69%|██████▊   | 10115/14725 [5:09:14<2:18:09,  1.
+0: {'loss': 0.7864, 'grad_norm': 0.7162669027766774, 'learning_rate': 2e-05, 'epoch': 0.69}
+0: 80s/it]
 69%|██████▊   | 10116/14725 [5:09:16<2:17:00,  1.78s/it]
 69%|██████▊   | 10117/14725 [5:09:17<2:17:46,  1.79s/it]
 69%|██████▊   | 10118/14725 [5:09:19<2:16:47,  1.78s/it]
 69%|██████▊   | 10119/14725 [5:09:21<2:17:03,  1.79s/it]
 69%|██████▊   | 10120/14725 [5:09:23<2:19:02,  1.81s/it]
                                                         

 69%|██████▊   | 10120/14725 [5:09:23<2:19:02,  1.81s/it]
 69%|██████▊   | 10121/14725 [5:09:25<2:20:01,  1.82s/it]
 69%|██████▊   | 10122/14725 [5:09:27<2:19:57,  1.82s/it]
 69%|██████▊   | 10123/14725 [5:09:28<2:21:14,  1.84s/it]
 69%|██████▉   | 10124/14725 [5:09:30<2:20:06,  1.83s/it]
 69%|██████▉   | 10125/14725 [5:09:32<2:19:43,  1.82s/it]
 69%|██████▉   | 10126/14725 [5:09:34<2:18:30,  1.81s/it]
 69%|██████▉   | 10127/14725 [5:09:36<2:17:13,  1.79s/it]
 69%|█████�
+0: {'loss': 0.8249, 'grad_norm': 0.6594496575611875, 'learning_rate': 2e-05, 'epoch': 0.69}
+0: ��▉   | 10128/14725 [5:09:37<2:16:58,  1.79s/it]
 69%|██████▉   | 10129/14725 [5:09:39<2:17:16,  1.79s/it]
 69%|██████▉   | 10130/14725 [5:09:41<2:18:45,  1.81s/it]
                                                         

 69%|██████▉   | 10130/14725 [5:09:41<2:18:45,  1.81s/it]
 69%|██████▉   | 10131/14725 [5:09:43<2:17:35,  1.80s/it]
 69%|██████▉   | 10132/14725 [5:09:45<2:16:42,  1.79s/it]
 69%|██████▉   | 10133/14725 [5:09:46<2:18:42,  1.81s/it]
 69%|██████▉   | 10134/14725 [5:09:48<2:18:33,  1.81s/it]
 69%|██████▉   | 10135/14725 [5:09:50<2:18:00,  1.80s/it]
 69%|██████▉   | 10136/14725 [5:09:52<2:17:29,  1.80s/it]
 69%|██████▉   | 10137/14725 [5:09:54<2:18:40,  1.81s/it]
 69%|██████▉   | 10138/14725 [5:09:56<2:22:10,  1.86s/it]
 69%|██████▉   | 10139/14725 [5:09:57<2:20:09,  1.83s/it]
 69%|██████▉   | 10140/14725 [5:09:5
+0: {'loss': 0.7998, 'grad_norm': 0.6846707946006588, 'learning_rate': 2e-05, 'epoch': 0.69}
+0: {'loss': 0.7838, 'grad_norm': 0.7520378346301114, 'learning_rate': 2e-05, 'epoch': 0.69}
+0: 9<2:18:58,  1.82s/it]
                                                         

 69%|██████▉   | 10140/14725 [5:09:59<2:18:58,  1.82s/it]
 69%|██████▉   | 10141/14725 [5:10:01<2:19:18,  1.82s/it]
 69%|██████▉   | 10142/14725 [5:10:03<2:18:28,  1.81s/it]
 69%|██████▉   | 10143/14725 [5:10:05<2:19:28,  1.83s/it]
 69%|██████▉   | 10144/14725 [5:10:06<2:17:53,  1.81s/it]
 69%|██████▉   | 10145/14725 [5:10:08<2:16:43,  1.79s/it]
 69%|██████▉   | 10146/14725 [5:10:10<2:16:01,  1.78s/it]
 69%|██████▉   | 10147/14725 [5:10:12<2:17:13,  1.80s/it]
 69%|██████▉   | 10148/14725 [5:10:14<2:16:40,  1.79s/it]
 69%|██████▉   | 10149/14725 [5:10:15<2:17:57,  1.81s/it]
 69%|██████▉   | 10150/14725 [5:10:17<2:17:41,  1.81s/it]
                                                         

 69%|██████▉   | 10150/14725 [5:10:17<2:17:41,  1.81s/it]
 69%|█████
+0: {'loss': 0.8237, 'grad_norm': 0.6826547116236076, 'learning_rate': 2e-05, 'epoch': 0.69}
+0: █▉   | 10151/14725 [5:10:19<2:17:47,  1.81s/it]
 69%|██████▉   | 10152/14725 [5:10:21<2:18:53,  1.82s/it]
 69%|██████▉   | 10153/14725 [5:10:23<2:18:33,  1.82s/it]
 69%|██████▉   | 10154/14725 [5:10:24<2:18:38,  1.82s/it]
 69%|██████▉   | 10155/14725 [5:10:26<2:16:59,  1.80s/it]
 69%|██████▉   | 10156/14725 [5:10:28<2:17:21,  1.80s/it]
 69%|██████▉   | 10157/14725 [5:10:30<2:18:22,  1.82s/it]
 69%|██████▉   | 10158/14725 [5:10:32<2:19:25,  1.83s/it]
 69%|██████▉   | 10159/14725 [5:10:34<2:18:58,  1.83s/it]
 69%|██████▉   | 10160/14725 [5:10:35<2:19:04,  1.83s/it]
                                                         

 69%|██████▉   | 10160/14725 [5:10:35<2:19:04,  1.83s/it]
 69%|██████▉   | 10161/14725 [5:10:37<2:19:10,  1.83s/it]
 69%|██████▉   | 10162/14725 [5:10:39<2:18:37,  1.82s/it]
 69%|██████▉   | 10163/14725 [5:10:
+0: {'loss': 0.8102, 'grad_norm': 0.7328697702697339, 'learning_rate': 2e-05, 'epoch': 0.69}
+0: 41<2:16:51,  1.80s/it]
 69%|██████▉   | 10164/14725 [5:10:43<2:15:58,  1.79s/it]
 69%|██████▉   | 10165/14725 [5:10:44<2:16:16,  1.79s/it]
 69%|██████▉   | 10166/14725 [5:10:46<2:18:35,  1.82s/it]
 69%|██████▉   | 10167/14725 [5:10:48<2:18:45,  1.83s/it]
 69%|██████▉   | 10168/14725 [5:10:50<2:18:18,  1.82s/it]
 69%|██████▉   | 10169/14725 [5:10:52<2:16:32,  1.80s/it]
 69%|██████▉   | 10170/14725 [5:10:54<2:17:44,  1.81s/it]
                                                         

 69%|██████▉   | 10170/14725 [5:10:54<2:17:44,  1.81s/it]
 69%|██████▉   | 10171/14725 [5:10:55<2:19:20,  1.84s/it]
 69%|██████▉   | 10172/14725 [5:10:57<2:17:16,  1.81s/it]
 69%|██████▉   | 10173/14725 [5:10:59<2:16:51,  1.80s/it]
 69%|██████▉   | 10174/14725 [5:11:01<2:17:56,  1.82s/it]
 69%|██████▉   | 10175/14725 [5:11:03<2:16:29,  1.80s/it]
 69%|�
+0: {'loss': 0.8085, 'grad_norm': 0.683529175412432, 'learning_rate': 2e-05, 'epoch': 0.69}
+0: ��█████▉   | 10176/14725 [5:11:04<2:16:04,  1.79s/it]
 69%|██████▉   | 10177/14725 [5:11:06<2:17:29,  1.81s/it]
 69%|██████▉   | 10178/14725 [5:11:08<2:15:52,  1.79s/it]
 69%|██████▉   | 10179/14725 [5:11:10<2:15:28,  1.79s/it]
 69%|██████▉   | 10180/14725 [5:11:11<2:15:14,  1.79s/it]
                                                         

 69%|██████▉   | 10180/14725 [5:11:11<2:15:14,  1.79s/it]
 69%|██████▉   | 10181/14725 [5:11:13<2:16:31,  1.80s/it]
 69%|██████▉   | 10182/14725 [5:11:15<2:16:27,  1.80s/it]
 69%|██████▉   | 10183/14725 [5:11:17<2:16:23,  1.80s/it]
 69%|██████▉   | 10184/14725 [5:11:19<2:16:16,  1.80s/it]
 69%|██████▉   | 10185/14725 [5:11:21<2:17:12,  1.81s/it]
 69%|██████▉   | 10186/14725 [5:11:22<2:18:14,  1.83s/it]
 69%|██████▉   | 10187/14725 [5:11:24<2:16:16,  1.80s/it]
 69%|██████▉   | 1018
+0: {'loss': 0.8126, 'grad_norm': 0.6659937734415334, 'learning_rate': 2e-05, 'epoch': 0.69}
+0: 8/14725 [5:11:26<2:16:41,  1.81s/it]
 69%|██████▉   | 10189/14725 [5:11:28<2:15:38,  1.79s/it]
 69%|██████▉   | 10190/14725 [5:11:30<2:16:59,  1.81s/it]
                                                         

 69%|██████▉   | 10190/14725 [5:11:30<2:16:59,  1.81s/it]
 69%|██████▉   | 10191/14725 [5:11:32<2:19:32,  1.85s/it]
 69%|██████▉   | 10192/14725 [5:11:33<2:18:56,  1.84s/it]
 69%|██████▉   | 10193/14725 [5:11:35<2:18:43,  1.84s/it]
 69%|██████▉   | 10194/14725 [5:11:37<2:17:35,  1.82s/it]
 69%|██████▉   | 10195/14725 [5:11:39<2:20:44,  1.86s/it]
 69%|██████▉   | 10196/14725 [5:11:41<2:18:35,  1.84s/it]
 69%|██████▉   | 10197/14725 [5:11:42<2:16:46,  1.81s/it]
 69%|██████▉   | 10198/14725 [5:11:44<2:17:15,  1.82s/it]
 69%|██████▉   | 10199/14725 [5:11:46<2:16:05,  1.80s/it]
 69%|██████▉   | 10200/14725 [5:11:48<2:17:23,  1.
+0: {'loss': 0.7982, 'grad_norm': 0.6394284415229422, 'learning_rate': 2e-05, 'epoch': 0.69}
+0: {'loss': 0.7924, 'grad_norm': 0.6665085876924557, 'learning_rate': 2e-05, 'epoch': 0.69}
+0: 82s/it]
                                                         

 69%|██████▉   | 10200/14725 [5:11:48<2:17:23,  1.82s/it]
 69%|██████▉   | 10201/14725 [5:11:50<2:17:15,  1.82s/it]
 69%|██████▉   | 10202/14725 [5:11:52<2:19:07,  1.85s/it]
 69%|██████▉   | 10203/14725 [5:11:53<2:17:28,  1.82s/it]
 69%|██████▉   | 10204/14725 [5:11:55<2:17:46,  1.83s/it]
 69%|██████▉   | 10205/14725 [5:11:57<2:17:08,  1.82s/it]
 69%|██████▉   | 10206/14725 [5:11:59<2:17:12,  1.82s/it]
 69%|██████▉   | 10207/14725 [5:12:01<2:16:03,  1.81s/it]
 69%|██████▉   | 10208/14725 [5:12:02<2:16:16,  1.81s/it]
 69%|██████▉   | 10209/14725 [5:12:04<2:15:29,  1.80s/it]
 69%|██████▉   | 10210/14725 [5:12:06<2:14:51,  1.79s/it]
                                                         

 69%|██████▉   | 10210/14725 [5:12:06<2:14:51,  1.79s/it]
 69%|██████▉   | 102
+0: {'loss': 0.8056, 'grad_norm': 0.7604219390670721, 'learning_rate': 2e-05, 'epoch': 0.69}
+0: 11/14725 [5:12:08<2:14:53,  1.79s/it]
 69%|██████▉   | 10212/14725 [5:12:10<2:14:41,  1.79s/it]
 69%|██████▉   | 10213/14725 [5:12:11<2:14:53,  1.79s/it]
 69%|██████▉   | 10214/14725 [5:12:13<2:17:52,  1.83s/it]
 69%|██████▉   | 10215/14725 [5:12:15<2:16:50,  1.82s/it]
 69%|██████▉   | 10216/14725 [5:12:17<2:16:12,  1.81s/it]
 69%|██████▉   | 10217/14725 [5:12:19<2:14:58,  1.80s/it]
 69%|██████▉   | 10218/14725 [5:12:20<2:15:10,  1.80s/it]
 69%|██████▉   | 10219/14725 [5:12:22<2:15:17,  1.80s/it]
 69%|██████▉   | 10220/14725 [5:12:24<2:16:33,  1.82s/it]
                                                         

 69%|██████▉   | 10220/14725 [5:12:24<2:16:33,  1.82s/it]
 69%|██████▉   | 10221/14725 [5:12:26<2:15:46,  1.81s/it]
 69%|██████▉   | 10222/14725 [5:12:28<2:18:09,  1.84s/it]
 69%|██████▉   | 10223/14725 [5:12:30<2:17:37,  1
+0: {'loss': 0.8171, 'grad_norm': 0.6800510498066545, 'learning_rate': 2e-05, 'epoch': 0.69}
+0: .83s/it]
 69%|██████▉   | 10224/14725 [5:12:31<2:15:31,  1.81s/it]
 69%|██████▉   | 10225/14725 [5:12:33<2:15:17,  1.80s/it]
 69%|██████▉   | 10226/14725 [5:12:35<2:17:39,  1.84s/it]
 69%|██████▉   | 10227/14725 [5:12:37<2:15:50,  1.81s/it]
 69%|██████▉   | 10228/14725 [5:12:39<2:15:01,  1.80s/it]
 69%|██████▉   | 10229/14725 [5:12:40<2:14:42,  1.80s/it]
 69%|██████▉   | 10230/14725 [5:12:42<2:13:51,  1.79s/it]
                                                         

 69%|██████▉   | 10230/14725 [5:12:42<2:13:51,  1.79s/it]
 69%|██████▉   | 10231/14725 [5:12:44<2:16:05,  1.82s/it]
 69%|██████▉   | 10232/14725 [5:12:46<2:15:10,  1.81s/it]
 69%|██████▉   | 10233/14725 [5:12:48<2:18:19,  1.85s/it]
 70%|██████▉   | 10234/14725 [5:12:50<2:17:48,  1.84s/it]
 70%|██████▉   | 10235/14725 [5:12:51<2:17:22,  1.84s/it]
 70%|█████
+0: {'loss': 0.7961, 'grad_norm': 0.6996349506957309, 'learning_rate': 2e-05, 'epoch': 0.7}
+0: █▉   | 10236/14725 [5:12:53<2:17:11,  1.83s/it]
 70%|██████▉   | 10237/14725 [5:12:55<2:15:00,  1.80s/it]
 70%|██████▉   | 10238/14725 [5:12:57<2:15:55,  1.82s/it]
 70%|██████▉   | 10239/14725 [5:12:59<2:15:27,  1.81s/it]
 70%|██████▉   | 10240/14725 [5:13:01<2:17:01,  1.83s/it]
                                                         

 70%|██████▉   | 10240/14725 [5:13:01<2:17:01,  1.83s/it]
 70%|██████▉   | 10241/14725 [5:13:02<2:16:13,  1.82s/it]
 70%|██████▉   | 10242/14725 [5:13:04<2:14:22,  1.80s/it]
 70%|██████▉   | 10243/14725 [5:13:06<2:14:47,  1.80s/it]
 70%|██████▉   | 10244/14725 [5:13:08<2:14:56,  1.81s/it]
 70%|██████▉   | 10245/14725 [5:13:10<2:14:25,  1.80s/it]
 70%|██████▉   | 10246/14725 [5:13:11<2:14:36,  1.80s/it]
 70%|██████▉   | 10247/14725 [5:13:13<2:14:46,  1.81s/it]
 70%|██████▉   | 10248/14725 [5:13:
+0: {'loss': 0.8051, 'grad_norm': 0.6608074707581021, 'learning_rate': 2e-05, 'epoch': 0.7}
+0: 15<2:14:33,  1.80s/it]
 70%|██████▉   | 10249/14725 [5:13:17<2:14:35,  1.80s/it]
 70%|██████▉   | 10250/14725 [5:13:19<2:13:45,  1.79s/it]
                                                         

 70%|██████▉   | 10250/14725 [5:13:19<2:13:45,  1.79s/it]
 70%|██████▉   | 10251/14725 [5:13:20<2:14:55,  1.81s/it]
 70%|██████▉   | 10252/14725 [5:13:22<2:14:17,  1.80s/it]
 70%|██████▉   | 10253/14725 [5:13:24<2:15:41,  1.82s/it]
 70%|██████▉   | 10254/14725 [5:13:26<2:14:59,  1.81s/it]
 70%|██████▉   | 10255/14725 [5:13:28<2:14:23,  1.80s/it]
 70%|██████▉   | 10256/14725 [5:13:29<2:16:02,  1.83s/it]
 70%|██████▉   | 10257/14725 [5:13:31<2:16:03,  1.83s/it]
 70%|██████▉   | 10258/14725 [5:13:33<2:15:58,  1.83s/it]
 70%|██████▉   | 10259/14725 [5:13:35<2:15:44,  1.82s/it]
 70%|██████▉   | 10260/14725 [5:13:37<2:14:34,  1.81s/it]
      
+0: {'loss': 0.8152, 'grad_norm': 0.6719501494670295, 'learning_rate': 2e-05, 'epoch': 0.7}
+0: {'loss': 0.8245, 'grad_norm': 0.6336335096159418, 'learning_rate': 2e-05, 'epoch': 0.7}
+0:                                                    

 70%|██████▉   | 10260/14725 [5:13:37<2:14:34,  1.81s/it]
 70%|██████▉   | 10261/14725 [5:13:39<2:14:45,  1.81s/it]
 70%|██████▉   | 10262/14725 [5:13:40<2:14:31,  1.81s/it]
 70%|██████▉   | 10263/14725 [5:13:42<2:13:52,  1.80s/it]
 70%|██████▉   | 10264/14725 [5:13:44<2:14:59,  1.82s/it]
 70%|██████▉   | 10265/14725 [5:13:46<2:14:32,  1.81s/it]
 70%|██████▉   | 10266/14725 [5:13:48<2:16:27,  1.84s/it]
 70%|██████▉   | 10267/14725 [5:13:49<2:15:02,  1.82s/it]
 70%|██████▉   | 10268/14725 [5:13:51<2:14:22,  1.81s/it]
 70%|██████▉   | 10269/14725 [5:13:53<2:14:46,  1.81s/it]
 70%|██████▉   | 10270/14725 [5:13:55<2:14:55,  1.82s/it]
                                                         

 70%|██████▉   | 10270/14725 [5:13:55<2:14:55,  1.82s/it]
 70%|██████▉   | 10271/14725 [5:13
+0: {'loss': 0.8108, 'grad_norm': 0.6468244843882668, 'learning_rate': 2e-05, 'epoch': 0.7}
+0: :57<2:14:47,  1.82s/it]
 70%|██████▉   | 10272/14725 [5:13:58<2:13:26,  1.80s/it]
 70%|██████▉   | 10273/14725 [5:14:00<2:16:16,  1.84s/it]
 70%|██████▉   | 10274/14725 [5:14:02<2:14:32,  1.81s/it]
 70%|██████▉   | 10275/14725 [5:14:04<2:13:23,  1.80s/it]
 70%|██████▉   | 10276/14725 [5:14:06<2:14:11,  1.81s/it]
 70%|██████▉   | 10277/14725 [5:14:07<2:13:04,  1.80s/it]
 70%|██████▉   | 10278/14725 [5:14:09<2:13:02,  1.79s/it]
 70%|██████▉   | 10279/14725 [5:14:11<2:11:59,  1.78s/it]
 70%|██████▉   | 10280/14725 [5:14:13<2:12:21,  1.79s/it]
                                                         

 70%|██████▉   | 10280/14725 [5:14:13<2:12:21,  1.79s/it]
 70%|██████▉   | 10281/14725 [5:14:15<2:13:06,  1.80s/it]
 70%|██████▉   | 10282/14725 [5:14:17<2:14:41,  1.82s/it]
 70%|██████▉   | 10283/14725 [5:14:18<2:14:02,  1.81s/it]
 70%|
+0: {'loss': 0.8072, 'grad_norm': 0.6732549161834401, 'learning_rate': 2e-05, 'epoch': 0.7}
+0: ██████▉   | 10284/14725 [5:14:20<2:14:00,  1.81s/it]
 70%|██████▉   | 10285/14725 [5:14:22<2:13:13,  1.80s/it]
 70%|██████▉   | 10286/14725 [5:14:24<2:16:03,  1.84s/it]
 70%|██████▉   | 10287/14725 [5:14:26<2:17:01,  1.85s/it]
 70%|██████▉   | 10288/14725 [5:14:27<2:15:05,  1.83s/it]
 70%|██████▉   | 10289/14725 [5:14:29<2:14:16,  1.82s/it]
 70%|██████▉   | 10290/14725 [5:14:31<2:13:50,  1.81s/it]
                                                         

 70%|██████▉   | 10290/14725 [5:14:31<2:13:50,  1.81s/it]
 70%|██████▉   | 10291/14725 [5:14:33<2:18:58,  1.88s/it]
 70%|██████▉   | 10292/14725 [5:14:35<2:15:48,  1.84s/it]
 70%|██████▉   | 10293/14725 [5:14:37<2:13:47,  1.81s/it]
 70%|██████▉   | 10294/14725 [5:14:38<2:13:26,  1.81s/it]
 70%|██████▉   | 10295/14725 [5:14:40<2:13:00,  1.80s/it]
 70%|██████▉   | 102
+0: {'loss': 0.803, 'grad_norm': 0.6480268629818942, 'learning_rate': 2e-05, 'epoch': 0.7}
+0: 96/14725 [5:14:42<2:14:30,  1.82s/it]
 70%|██████▉   | 10297/14725 [5:14:44<2:13:38,  1.81s/it]
 70%|██████▉   | 10298/14725 [5:14:46<2:11:57,  1.79s/it]
 70%|██████▉   | 10299/14725 [5:14:47<2:11:39,  1.78s/it]
 70%|██████▉   | 10300/14725 [5:14:49<2:11:38,  1.78s/it]
                                                         

 70%|██████▉   | 10300/14725 [5:14:49<2:11:38,  1.78s/it]
 70%|██████▉   | 10301/14725 [5:14:51<2:12:58,  1.80s/it]
 70%|██████▉   | 10302/14725 [5:14:53<2:15:42,  1.84s/it]
 70%|██████▉   | 10303/14725 [5:14:55<2:15:30,  1.84s/it]
 70%|██████▉   | 10304/14725 [5:14:56<2:13:03,  1.81s/it]
 70%|██████▉   | 10305/14725 [5:14:58<2:12:42,  1.80s/it]
 70%|██████▉   | 10306/14725 [5:15:00<2:13:04,  1.81s/it]
 70%|██████▉   | 10307/14725 [5:15:02<2:11:58,  1.79s/it]
 70%|███████   | 10308/14725 [5:15:04<2:13:53,  1
+0: {'loss': 0.8046, 'grad_norm': 0.6604336469990078, 'learning_rate': 2e-05, 'epoch': 0.7}
+0: .82s/it]
 70%|███████   | 10309/14725 [5:15:06<2:13:39,  1.82s/it]
 70%|███████   | 10310/14725 [5:15:07<2:12:29,  1.80s/it]
                                                         

 70%|███████   | 10310/14725 [5:15:07<2:12:29,  1.80s/it]
 70%|███████   | 10311/14725 [5:15:09<2:11:42,  1.79s/it]
 70%|███████   | 10312/14725 [5:15:11<2:13:20,  1.81s/it]
 70%|███████   | 10313/14725 [5:15:13<2:14:09,  1.82s/it]
 70%|███████   | 10314/14725 [5:15:15<2:13:54,  1.82s/it]
 70%|███████   | 10315/14725 [5:15:16<2:13:30,  1.82s/it]
 70%|███████   | 10316/14725 [5:15:18<2:13:14,  1.81s/it]
 70%|███████   | 10317/14725 [5:15:20<2:13:58,  1.82s/it]
 70%|███████   | 10318/14725 [5:15:22<2:14:24,  1.83s/it]
 70%|███████   | 10319/14725 [5:15:24<2:14:55,  1.84s/it]
 70%|███████   | 10320/14725 [5:15:26<2:14:34,  1.83s/it]
                    
+0: {'loss': 0.8031, 'grad_norm': 0.6703215118212256, 'learning_rate': 2e-05, 'epoch': 0.7}
+0: {'loss': 0.7999, 'grad_norm': 0.6878806233748783, 'learning_rate': 2e-05, 'epoch': 0.7}
+0:                                      

 70%|███████   | 10320/14725 [5:15:26<2:14:34,  1.83s/it]
 70%|███████   | 10321/14725 [5:15:27<2:15:02,  1.84s/it]
 70%|███████   | 10322/14725 [5:15:29<2:14:47,  1.84s/it]
 70%|███████   | 10323/14725 [5:15:31<2:15:29,  1.85s/it]
 70%|███████   | 10324/14725 [5:15:33<2:16:11,  1.86s/it]
 70%|███████   | 10325/14725 [5:15:35<2:15:52,  1.85s/it]
 70%|███████   | 10326/14725 [5:15:37<2:13:57,  1.83s/it]
 70%|███████   | 10327/14725 [5:15:38<2:14:42,  1.84s/it]
 70%|███████   | 10328/14725 [5:15:40<2:14:45,  1.84s/it]
 70%|███████   | 10329/14725 [5:15:42<2:14:16,  1.83s/it]
 70%|███████   | 10330/14725 [5:15:44<2:13:39,  1.82s/it]
                                                         

 70%|███████   | 10330/14725 [5:15:44<2:13:39,  1.82s/it]
 70%|███████   | 10331/14725 [5:15:46<2:13:21,  
+0: {'loss': 0.7939, 'grad_norm': 0.7216599130651096, 'learning_rate': 2e-05, 'epoch': 0.7}
+0: 1.82s/it]
 70%|███████   | 10332/14725 [5:15:48<2:12:31,  1.81s/it]
 70%|███████   | 10333/14725 [5:15:49<2:10:58,  1.79s/it]
 70%|███████   | 10334/14725 [5:15:51<2:10:28,  1.78s/it]
 70%|███████   | 10335/14725 [5:15:53<2:12:20,  1.81s/it]
 70%|███████   | 10336/14725 [5:15:55<2:12:18,  1.81s/it]
 70%|███████   | 10337/14725 [5:15:57<2:12:01,  1.81s/it]
 70%|███████   | 10338/14725 [5:15:58<2:13:22,  1.82s/it]
 70%|███████   | 10339/14725 [5:16:00<2:12:44,  1.82s/it]
 70%|███████   | 10340/14725 [5:16:02<2:12:25,  1.81s/it]
                                                         

 70%|███████   | 10340/14725 [5:16:02<2:12:25,  1.81s/it]
 70%|███████   | 10341/14725 [5:16:04<2:15:47,  1.86s/it]
 70%|███████   | 10342/14725 [5:16:06<2:13:58,  1.83s/it]
 70%|███████   | 10343/14725 [5:16:08<2:13:39,  1.83s/it]
 70%|████�
+0: {'loss': 0.8008, 'grad_norm': 0.7143927066194773, 'learning_rate': 2e-05, 'epoch': 0.7}
+0: �██   | 10344/14725 [5:16:09<2:13:52,  1.83s/it]
 70%|███████   | 10345/14725 [5:16:11<2:12:06,  1.81s/it]
 70%|███████   | 10346/14725 [5:16:13<2:11:26,  1.80s/it]
 70%|███████   | 10347/14725 [5:16:15<2:12:03,  1.81s/it]
 70%|███████   | 10348/14725 [5:16:17<2:12:25,  1.82s/it]
 70%|███████   | 10349/14725 [5:16:18<2:12:06,  1.81s/it]
 70%|███████   | 10350/14725 [5:16:20<2:12:48,  1.82s/it]
                                                         

 70%|███████   | 10350/14725 [5:16:20<2:12:48,  1.82s/it]
 70%|███████   | 10351/14725 [5:16:22<2:13:41,  1.83s/it]
 70%|███████   | 10352/14725 [5:16:24<2:11:36,  1.81s/it]
 70%|███████   | 10353/14725 [5:16:26<2:10:32,  1.79s/it]
 70%|███████   | 10354/14725 [5:16:27<2:10:49,  1.80s/it]
 70%|███████   | 10355/14725 [5:16:29<2:12:13,  1.82s/it]
 70%|███████   | 10356/14725 [5:16
+0: {'loss': 0.7898, 'grad_norm': 0.6940023706360505, 'learning_rate': 2e-05, 'epoch': 0.7}
+0: :31<2:12:48,  1.82s/it]
 70%|███████   | 10357/14725 [5:16:33<2:11:24,  1.81s/it]
 70%|███████   | 10358/14725 [5:16:35<2:11:22,  1.80s/it]
 70%|███████   | 10359/14725 [5:16:36<2:10:27,  1.79s/it]
 70%|███████   | 10360/14725 [5:16:38<2:12:01,  1.81s/it]
                                                         

 70%|███████   | 10360/14725 [5:16:38<2:12:01,  1.81s/it]
 70%|███████   | 10361/14725 [5:16:40<2:12:22,  1.82s/it]
 70%|███████   | 10362/14725 [5:16:42<2:12:35,  1.82s/it]
 70%|███████   | 10363/14725 [5:16:44<2:11:22,  1.81s/it]
 70%|███████   | 10364/14725 [5:16:46<2:10:16,  1.79s/it]
 70%|███████   | 10365/14725 [5:16:47<2:09:07,  1.78s/it]
 70%|███████   | 10366/14725 [5:16:49<2:08:39,  1.77s/it]
 70%|███████   | 10367/14725 [5:16:51<2:09:43,  1.79s/it]
 70%|███████   | 10368/14725 [5:16:53<2:08:43,  1.77s/it]
 70%|
+0: {'loss': 0.8157, 'grad_norm': 0.722971337419222, 'learning_rate': 2e-05, 'epoch': 0.7}
+0: ███████   | 10369/14725 [5:16:54<2:10:47,  1.80s/it]
 70%|███████   | 10370/14725 [5:16:56<2:10:08,  1.79s/it]
                                                         

 70%|███████   | 10370/14725 [5:16:56<2:10:08,  1.79s/it]
 70%|███████   | 10371/14725 [5:16:58<2:09:52,  1.79s/it]
 70%|███████   | 10372/14725 [5:17:00<2:11:20,  1.81s/it]
 70%|███████   | 10373/14725 [5:17:02<2:14:59,  1.86s/it]
 70%|███████   | 10374/14725 [5:17:04<2:14:24,  1.85s/it]
 70%|███████   | 10375/14725 [5:17:05<2:13:00,  1.83s/it]
 70%|███████   | 10376/14725 [5:17:07<2:11:43,  1.82s/it]
 70%|███████   | 10377/14725 [5:17:09<2:11:41,  1.82s/it]
 70%|███████   | 10378/14725 [5:17:11<2:12:01,  1.82s/it]
 70%|███████   | 10379/14725 [5:17:13<2:11:36,  1.82s/it]
 70%|███████   | 10380/14725 [5:17:15<2:11:41,  1.82s/it]
                                  
+0: {'loss': 0.8289, 'grad_norm': 0.7259410073689115, 'learning_rate': 2e-05, 'epoch': 0.7}
+0: {'loss': 0.7852, 'grad_norm': 0.6860808809490536, 'learning_rate': 2e-05, 'epoch': 0.71}
+0:                        

 70%|███████   | 10380/14725 [5:17:15<2:11:41,  1.82s/it]
 70%|███████   | 10381/14725 [5:17:16<2:10:24,  1.80s/it]
 71%|███████   | 10382/14725 [5:17:18<2:09:35,  1.79s/it]
 71%|███████   | 10383/14725 [5:17:20<2:08:29,  1.78s/it]
 71%|███████   | 10384/14725 [5:17:22<2:08:12,  1.77s/it]
 71%|███████   | 10385/14725 [5:17:23<2:08:48,  1.78s/it]
 71%|███████   | 10386/14725 [5:17:25<2:08:33,  1.78s/it]
 71%|███████   | 10387/14725 [5:17:27<2:10:38,  1.81s/it]
 71%|███████   | 10388/14725 [5:17:29<2:09:30,  1.79s/it]
 71%|███████   | 10389/14725 [5:17:31<2:12:27,  1.83s/it]
 71%|███████   | 10390/14725 [5:17:32<2:11:22,  1.82s/it]
                                                         

 71%|███████   | 10390/14725 [5:17:32<2:11:22,  1.82s/it]
 71%|███████   | 10391/14725 [5:17:34<2:10:15,  1.80s/it]
 71%
+0: {'loss': 0.7939, 'grad_norm': 0.708923679920755, 'learning_rate': 2e-05, 'epoch': 0.71}
+0: |███████   | 10392/14725 [5:17:36<2:10:15,  1.80s/it]
 71%|███████   | 10393/14725 [5:17:38<2:11:57,  1.83s/it]
 71%|███████   | 10394/14725 [5:17:40<2:11:14,  1.82s/it]
 71%|███████   | 10395/14725 [5:17:42<2:10:45,  1.81s/it]
 71%|███████   | 10396/14725 [5:17:43<2:11:34,  1.82s/it]
 71%|███████   | 10397/14725 [5:17:45<2:11:05,  1.82s/it]
 71%|███████   | 10398/14725 [5:17:47<2:10:12,  1.81s/it]
 71%|███████   | 10399/14725 [5:17:49<2:11:37,  1.83s/it]
 71%|███████   | 10400/14725 [5:17:51<2:09:57,  1.80s/it]
                                                         

 71%|███████   | 10400/14725 [5:17:51<2:09:57,  1.80s/it]
 71%|███████   | 10401/14725 [5:17:52<2:09:23,  1.80s/it]
 71%|███████   | 10402/14725 [5:17:54<2:10:06,  1.81s/it]
 71%|███████   | 10403/14725 [5:17:56<2:10:58,  1.82s/it]
 71%|███████   | 10
+0: {'loss': 0.8126, 'grad_norm': 0.6918595242138947, 'learning_rate': 2e-05, 'epoch': 0.71}
+0: 404/14725 [5:17:58<2:09:46,  1.80s/it]
 71%|███████   | 10405/14725 [5:18:00<2:08:54,  1.79s/it]
 71%|███████   | 10406/14725 [5:18:01<2:09:13,  1.80s/it]
 71%|███████   | 10407/14725 [5:18:03<2:08:53,  1.79s/it]
 71%|███████   | 10408/14725 [5:18:05<2:11:16,  1.82s/it]
 71%|███████   | 10409/14725 [5:18:07<2:24:19,  2.01s/it]
 71%|███████   | 10410/14725 [5:18:09<2:19:29,  1.94s/it]
                                                         

 71%|███████   | 10410/14725 [5:18:09<2:19:29,  1.94s/it]
 71%|███████   | 10411/14725 [5:18:11<2:16:28,  1.90s/it]
 71%|███████   | 10412/14725 [5:18:13<2:14:42,  1.87s/it]
 71%|███████   | 10413/14725 [5:18:15<2:11:53,  1.84s/it]
 71%|███████   | 10414/14725 [5:18:16<2:10:14,  1.81s/it]
 71%|███████   | 10415/14725 [5:18:18<2:10:05,  1.81s/it]
 71%|███████   | 10416/14725 [5:18:20<2:10:11,  
+0: {'loss': 0.8263, 'grad_norm': 0.6786710165597336, 'learning_rate': 2e-05, 'epoch': 0.71}
+0: 1.81s/it]
 71%|███████   | 10417/14725 [5:18:22<2:13:14,  1.86s/it]
 71%|███████   | 10418/14725 [5:18:24<2:10:39,  1.82s/it]
 71%|███████   | 10419/14725 [5:18:26<2:10:58,  1.82s/it]
 71%|███████   | 10420/14725 [5:18:27<2:09:56,  1.81s/it]
                                                         

 71%|███████   | 10420/14725 [5:18:27<2:09:56,  1.81s/it]
 71%|███████   | 10421/14725 [5:18:29<2:08:38,  1.79s/it]
 71%|███████   | 10422/14725 [5:18:31<2:16:57,  1.91s/it]
 71%|███████   | 10423/14725 [5:18:33<2:14:02,  1.87s/it]
 71%|███████   | 10424/14725 [5:18:35<2:12:16,  1.85s/it]
 71%|███████   | 10425/14725 [5:18:37<2:13:36,  1.86s/it]
 71%|███████   | 10426/14725 [5:18:38<2:11:46,  1.84s/it]
 71%|███████   | 10427/14725 [5:18:40<2:11:42,  1.84s/it]
 71%|███████   | 10428/14725 [5:18:42<2:15:11,  1.89s/it]
 71%|████�
+0: {'loss': 0.8023, 'grad_norm': 0.7555946165810048, 'learning_rate': 2e-05, 'epoch': 0.71}
+0: �██   | 10429/14725 [5:18:44<2:12:44,  1.85s/it]
 71%|███████   | 10430/14725 [5:18:46<2:11:51,  1.84s/it]
                                                         

 71%|███████   | 10430/14725 [5:18:46<2:11:51,  1.84s/it]
 71%|███████   | 10431/14725 [5:18:48<2:13:07,  1.86s/it]
 71%|███████   | 10432/14725 [5:18:50<2:11:49,  1.84s/it]
 71%|███████   | 10433/14725 [5:18:51<2:11:46,  1.84s/it]
 71%|███████   | 10434/14725 [5:18:53<2:10:35,  1.83s/it]
 71%|███████   | 10435/14725 [5:18:55<2:11:15,  1.84s/it]
 71%|███████   | 10436/14725 [5:18:57<2:11:54,  1.85s/it]
 71%|███████   | 10437/14725 [5:18:59<2:11:03,  1.83s/it]
 71%|███████   | 10438/14725 [5:19:01<2:15:29,  1.90s/it]
 71%|███████   | 10439/14725 [5:19:03<2:14:28,  1.88s/it]
 71%|███████   | 10440/14725 [5:19:04<2:12:12,  1.85s/it]
                                                
+0: {'loss': 0.7968, 'grad_norm': 0.7006976997731018, 'learning_rate': 2e-05, 'epoch': 0.71}
+0: {'loss': 0.805, 'grad_norm': 0.6812600034941596, 'learning_rate': 2e-05, 'epoch': 0.71}
+0:          

 71%|███████   | 10440/14725 [5:19:04<2:12:12,  1.85s/it]
 71%|███████   | 10441/14725 [5:19:07<2:19:55,  1.96s/it]
 71%|███████   | 10442/14725 [5:19:08<2:16:03,  1.91s/it]
 71%|███████   | 10443/14725 [5:19:10<2:12:47,  1.86s/it]
 71%|███████   | 10444/14725 [5:19:12<2:11:12,  1.84s/it]
 71%|███████   | 10445/14725 [5:19:14<2:10:35,  1.83s/it]
 71%|███████   | 10446/14725 [5:19:16<2:10:50,  1.83s/it]
 71%|███████   | 10447/14725 [5:19:17<2:09:42,  1.82s/it]
 71%|███████   | 10448/14725 [5:19:19<2:08:03,  1.80s/it]
 71%|███████   | 10449/14725 [5:19:21<2:07:06,  1.78s/it]
 71%|███████   | 10450/14725 [5:19:23<2:08:18,  1.80s/it]
                                                         

 71%|███████   | 10450/14725 [5:19:23<2:08:18,  1.80s/it]
 71%|███████   | 10451/14725 [5:19:25<2:08:31,  1.80s/it]
 71%|████�
+0: {'loss': 0.7869, 'grad_norm': 0.675020260604307, 'learning_rate': 2e-05, 'epoch': 0.71}
+0: ��██   | 10452/14725 [5:19:26<2:07:34,  1.79s/it]
 71%|███████   | 10453/14725 [5:19:28<2:10:42,  1.84s/it]
 71%|███████   | 10454/14725 [5:19:30<2:10:46,  1.84s/it]
 71%|███████   | 10455/14725 [5:19:32<2:08:40,  1.81s/it]
 71%|███████   | 10456/14725 [5:19:34<2:07:17,  1.79s/it]
 71%|███████   | 10457/14725 [5:19:35<2:07:14,  1.79s/it]
 71%|███████   | 10458/14725 [5:19:37<2:08:57,  1.81s/it]
 71%|███████   | 10459/14725 [5:19:39<2:16:20,  1.92s/it]
 71%|███████   | 10460/14725 [5:19:41<2:13:02,  1.87s/it]
                                                         

 71%|███████   | 10460/14725 [5:19:41<2:13:02,  1.87s/it]
 71%|███████   | 10461/14725 [5:19:43<2:12:37,  1.87s/it]
 71%|███████   | 10462/14725 [5:19:45<2:10:26,  1.84s/it]
 71%|███████   | 10463/14725 [5:19:47<2:10:23,  1.84s/it]
 71%|███████   | 10464/14725 [5:1
+0: {'loss': 0.791, 'grad_norm': 0.7326913868675499, 'learning_rate': 2e-05, 'epoch': 0.71}
+0: 9:48<2:09:34,  1.82s/it]
 71%|███████   | 10465/14725 [5:19:50<2:09:38,  1.83s/it]
 71%|███████   | 10466/14725 [5:19:52<2:08:02,  1.80s/it]
 71%|███████   | 10467/14725 [5:19:54<2:07:03,  1.79s/it]
 71%|███████   | 10468/14725 [5:19:56<2:08:26,  1.81s/it]
 71%|███████   | 10469/14725 [5:19:57<2:07:30,  1.80s/it]
 71%|███████   | 10470/14725 [5:19:59<2:09:21,  1.82s/it]
                                                         

 71%|███████   | 10470/14725 [5:19:59<2:09:21,  1.82s/it]
 71%|███████   | 10471/14725 [5:20:01<2:08:13,  1.81s/it]
 71%|███████   | 10472/14725 [5:20:03<2:09:18,  1.82s/it]
 71%|███████   | 10473/14725 [5:20:05<2:07:44,  1.80s/it]
 71%|███████   | 10474/14725 [5:20:07<2:08:56,  1.82s/it]
 71%|███████   | 10475/14725 [5:20:08<2:09:16,  1.83s/it]
 71%|███████   | 10476/14725 [5:20:10<2:07:47,  1.80s/it]
 71%
+0: {'loss': 0.8051, 'grad_norm': 0.7315894229267166, 'learning_rate': 2e-05, 'epoch': 0.71}
+0: |███████   | 10477/14725 [5:20:12<2:07:03,  1.79s/it]
 71%|███████   | 10478/14725 [5:20:14<2:07:35,  1.80s/it]
 71%|███████   | 10479/14725 [5:20:16<2:08:04,  1.81s/it]
 71%|███████   | 10480/14725 [5:20:17<2:07:49,  1.81s/it]
                                                         

 71%|███████   | 10480/14725 [5:20:17<2:07:49,  1.81s/it]
 71%|███████   | 10481/14725 [5:20:19<2:06:28,  1.79s/it]
 71%|███████   | 10482/14725 [5:20:21<2:06:34,  1.79s/it]
 71%|███████   | 10483/14725 [5:20:23<2:06:52,  1.79s/it]
 71%|███████   | 10484/14725 [5:20:24<2:06:50,  1.79s/it]
 71%|███████   | 10485/14725 [5:20:26<2:05:55,  1.78s/it]
 71%|███████   | 10486/14725 [5:20:28<2:05:53,  1.78s/it]
 71%|███████   | 10487/14725 [5:20:30<2:05:58,  1.78s/it]
 71%|███████   | 10488/14725 [5:20:32<2:05:26,  1.78s/it]
 71%|███████   | 10
+0: {'loss': 0.8148, 'grad_norm': 0.7059893278683347, 'learning_rate': 2e-05, 'epoch': 0.71}
+0: 489/14725 [5:20:33<2:05:30,  1.78s/it]
 71%|███████   | 10490/14725 [5:20:35<2:07:13,  1.80s/it]
                                                         

 71%|███████   | 10490/14725 [5:20:35<2:07:13,  1.80s/it]
 71%|███████   | 10491/14725 [5:20:37<2:07:01,  1.80s/it]
 71%|███████▏  | 10492/14725 [5:20:39<2:06:16,  1.79s/it]
 71%|███████▏  | 10493/14725 [5:20:41<2:06:29,  1.79s/it]
 71%|███████▏  | 10494/14725 [5:20:42<2:05:57,  1.79s/it]
 71%|███████▏  | 10495/14725 [5:20:44<2:06:09,  1.79s/it]
 71%|███████▏  | 10496/14725 [5:20:46<2:06:00,  1.79s/it]
 71%|███████▏  | 10497/14725 [5:20:48<2:07:11,  1.81s/it]
 71%|███████▏  | 10498/14725 [5:20:50<2:08:06,  1.82s/it]
 71%|███████▏  | 10499/14725 [5:20:51<2:08:06,  1.82s/it]
 71%|███████▏  | 10500/14725 [5:20:53<2:06:28,  1.80s/it]
                                            
+0: {'loss': 0.809, 'grad_norm': 0.6624717590141685, 'learning_rate': 2e-05, 'epoch': 0.71}
+0: {'loss': 0.8233, 'grad_norm': 0.72186248281183, 'learning_rate': 2e-05, 'epoch': 0.71}
+0:              

 71%|███████▏  | 10500/14725 [5:20:53<2:06:28,  1.80s/it]
 71%|███████▏  | 10501/14725 [5:20:55<2:06:01,  1.79s/it]
 71%|███████▏  | 10502/14725 [5:20:57<2:07:17,  1.81s/it]
 71%|███████▏  | 10503/14725 [5:20:59<2:06:11,  1.79s/it]
 71%|███████▏  | 10504/14725 [5:21:00<2:05:57,  1.79s/it]
 71%|███████▏  | 10505/14725 [5:21:02<2:05:20,  1.78s/it]
 71%|███████▏  | 10506/14725 [5:21:04<2:05:15,  1.78s/it]
 71%|███████▏  | 10507/14725 [5:21:06<2:04:50,  1.78s/it]
 71%|███████▏  | 10508/14725 [5:21:07<2:04:48,  1.78s/it]
 71%|███████▏  | 10509/14725 [5:21:09<2:06:26,  1.80s/it]
 71%|███████▏  | 10510/14725 [5:21:11<2:05:44,  1.79s/it]
                                                         

 71%|███████▏  | 10510/14725 [5:21:11<2:05:44,  1.79s/it]
 71%|███████▏  | 10511/14725 [5:21:13<2:04:49,
+0: {'loss': 0.8194, 'grad_norm': 0.7586663119679491, 'learning_rate': 2e-05, 'epoch': 0.71}
+0:   1.78s/it]
 71%|███████▏  | 10512/14725 [5:21:15<2:05:33,  1.79s/it]
 71%|███████▏  | 10513/14725 [5:21:16<2:04:48,  1.78s/it]
 71%|███████▏  | 10514/14725 [5:21:18<2:05:31,  1.79s/it]
 71%|███████▏  | 10515/14725 [5:21:20<2:06:22,  1.80s/it]
 71%|███████▏  | 10516/14725 [5:21:22<2:05:51,  1.79s/it]
 71%|███████▏  | 10517/14725 [5:21:24<2:05:29,  1.79s/it]
 71%|███████▏  | 10518/14725 [5:21:25<2:04:38,  1.78s/it]
 71%|███████▏  | 10519/14725 [5:21:27<2:06:05,  1.80s/it]
 71%|███████▏  | 10520/14725 [5:21:29<2:06:41,  1.81s/it]
                                                         

 71%|███████▏  | 10520/14725 [5:21:29<2:06:41,  1.81s/it]
 71%|███████▏  | 10521/14725 [5:21:31<2:08:17,  1.83s/it]
 71%|███████▏  | 10522/14725 [5:21:33<2:06:39,  1.81s/it]
 71%|███████▏  | 10523/14725 [5:21:34<2:05:43,  1
+0: {'loss': 0.795, 'grad_norm': 0.704736958558599, 'learning_rate': 2e-05, 'epoch': 0.72}
+0: .80s/it]
 71%|███████▏  | 10524/14725 [5:21:36<2:05:15,  1.79s/it]
 71%|███████▏  | 10525/14725 [5:21:38<2:04:43,  1.78s/it]
 71%|███████▏  | 10526/14725 [5:21:40<2:05:34,  1.79s/it]
 71%|███████▏  | 10527/14725 [5:21:42<2:05:08,  1.79s/it]
 71%|███████▏  | 10528/14725 [5:21:43<2:05:18,  1.79s/it]
 72%|███████▏  | 10529/14725 [5:21:45<2:04:32,  1.78s/it]
 72%|███████▏  | 10530/14725 [5:21:47<2:03:58,  1.77s/it]
                                                         

 72%|███████▏  | 10530/14725 [5:21:47<2:03:58,  1.77s/it]
 72%|███████▏  | 10531/14725 [5:21:49<2:05:34,  1.80s/it]
 72%|███████▏  | 10532/14725 [5:21:51<2:06:54,  1.82s/it]
 72%|███████▏  | 10533/14725 [5:21:52<2:05:49,  1.80s/it]
 72%|███████▏  | 10534/14725 [5:21:54<2:05:51,  1.80s/it]
 72%|███████▏  | 10535/14725 [5:21:56<2:07:50,  1.83
+0: {'loss': 0.8329, 'grad_norm': 0.6785610630816337, 'learning_rate': 2e-05, 'epoch': 0.72}
+0: s/it]
 72%|███████▏  | 10536/14725 [5:21:58<2:08:07,  1.84s/it]
 72%|███████▏  | 10537/14725 [5:22:00<2:07:32,  1.83s/it]
 72%|███████▏  | 10538/14725 [5:22:01<2:06:46,  1.82s/it]
 72%|███████▏  | 10539/14725 [5:22:03<2:06:23,  1.81s/it]
 72%|███████▏  | 10540/14725 [5:22:05<2:04:38,  1.79s/it]
                                                         

 72%|███████▏  | 10540/14725 [5:22:05<2:04:38,  1.79s/it]
 72%|███████▏  | 10541/14725 [5:22:07<2:04:56,  1.79s/it]
 72%|███████▏  | 10542/14725 [5:22:09<2:05:17,  1.80s/it]
 72%|███████▏  | 10543/14725 [5:22:10<2:05:10,  1.80s/it]
 72%|███████▏  | 10544/14725 [5:22:12<2:06:09,  1.81s/it]
 72%|███████▏  | 10545/14725 [5:22:14<2:04:42,  1.79s/it]
 72%|███████▏  | 10546/14725 [5:22:16<2:06:35,  1.82s/it]
 72%|███████▏  | 10547/14725 [5:22:18<2:05:34,  1.80s/i
+0: {'loss': 0.8208, 'grad_norm': 0.6925101492533606, 'learning_rate': 2e-05, 'epoch': 0.72}
+0: t]
 72%|███████▏  | 10548/14725 [5:22:19<2:06:15,  1.81s/it]
 72%|███████▏  | 10549/14725 [5:22:21<2:05:47,  1.81s/it]
 72%|███████▏  | 10550/14725 [5:22:23<2:05:57,  1.81s/it]
                                                         

 72%|███████▏  | 10550/14725 [5:22:23<2:05:57,  1.81s/it]
 72%|███████▏  | 10551/14725 [5:22:25<2:05:36,  1.81s/it]
 72%|███████▏  | 10552/14725 [5:22:27<2:05:38,  1.81s/it]
 72%|███████▏  | 10553/14725 [5:22:28<2:04:29,  1.79s/it]
 72%|███████▏  | 10554/14725 [5:22:30<2:04:06,  1.79s/it]
 72%|███████▏  | 10555/14725 [5:22:32<2:04:57,  1.80s/it]
 72%|███████▏  | 10556/14725 [5:22:34<2:04:48,  1.80s/it]
 72%|███████▏  | 10557/14725 [5:22:36<2:05:13,  1.80s/it]
 72%|███████▏  | 10558/14725 [5:22:38<2:06:01,  1.81s/it]
 72%|███████▏  | 10559/14725 [5:22:39<2:04:17,  1.79s/it]
+0: {'loss': 0.7949, 'grad_norm': 0.6604259323706766, 'learning_rate': 2e-05, 'epoch': 0.72}
+0: {'loss': 0.7892, 'grad_norm': 0.6998043334795794, 'learning_rate': 2e-05, 'epoch': 0.72}
+0:  72%|███████▏  | 10560/14725 [5:22:41<2:05:05,  1.80s/it]
                                                         

 72%|███████▏  | 10560/14725 [5:22:41<2:05:05,  1.80s/it]
 72%|███████▏  | 10561/14725 [5:22:43<2:04:46,  1.80s/it]
 72%|███████▏  | 10562/14725 [5:22:45<2:05:25,  1.81s/it]
 72%|███████▏  | 10563/14725 [5:22:47<2:05:52,  1.81s/it]
 72%|███████▏  | 10564/14725 [5:22:48<2:05:53,  1.82s/it]
 72%|███████▏  | 10565/14725 [5:22:50<2:06:03,  1.82s/it]
 72%|███████▏  | 10566/14725 [5:22:52<2:06:21,  1.82s/it]
 72%|███████▏  | 10567/14725 [5:22:54<2:04:37,  1.80s/it]
 72%|███████▏  | 10568/14725 [5:22:56<2:05:07,  1.81s/it]
 72%|███████▏  | 10569/14725 [5:22:57<2:05:35,  1.81s/it]
 72%|███████▏  | 10570/14725 [5:22:59<2:04:11,  1.79s/it]
                                                         

 72%|████�
+0: {'loss': 0.8094, 'grad_norm': 0.6821137452429934, 'learning_rate': 2e-05, 'epoch': 0.72}
+0: ��██▏  | 10570/14725 [5:22:59<2:04:11,  1.79s/it]
 72%|███████▏  | 10571/14725 [5:23:01<2:05:59,  1.82s/it]
 72%|███████▏  | 10572/14725 [5:23:03<2:05:58,  1.82s/it]
 72%|███████▏  | 10573/14725 [5:23:05<2:06:59,  1.84s/it]
 72%|███████▏  | 10574/14725 [5:23:07<2:05:57,  1.82s/it]
 72%|███████▏  | 10575/14725 [5:23:08<2:05:40,  1.82s/it]
 72%|███████▏  | 10576/14725 [5:23:10<2:04:51,  1.81s/it]
 72%|███████▏  | 10577/14725 [5:23:12<2:04:56,  1.81s/it]
 72%|███████▏  | 10578/14725 [5:23:14<2:07:15,  1.84s/it]
 72%|███████▏  | 10579/14725 [5:23:16<2:06:03,  1.82s/it]
 72%|███████▏  | 10580/14725 [5:23:17<2:06:00,  1.82s/it]
                                                         

 72%|███████▏  | 10580/14725 [5:23:17<2:06:00,  1.82s/it]
 72%|███████▏  | 10581/14725 [5:23:19<2:06:02,  1.82s/it]
 72%|█████�
+0: {'loss': 0.8028, 'grad_norm': 0.6821780283540493, 'learning_rate': 2e-05, 'epoch': 0.72}
+0: ��█▏  | 10582/14725 [5:23:21<2:06:05,  1.83s/it]
 72%|███████▏  | 10583/14725 [5:23:23<2:05:17,  1.81s/it]
 72%|███████▏  | 10584/14725 [5:23:25<2:05:36,  1.82s/it]
 72%|███████▏  | 10585/14725 [5:23:27<2:06:01,  1.83s/it]
 72%|███████▏  | 10586/14725 [5:23:28<2:05:08,  1.81s/it]
 72%|███████▏  | 10587/14725 [5:23:30<2:04:25,  1.80s/it]
 72%|███████▏  | 10588/14725 [5:23:32<2:04:23,  1.80s/it]
 72%|███████▏  | 10589/14725 [5:23:34<2:05:30,  1.82s/it]
 72%|███████▏  | 10590/14725 [5:23:36<2:07:35,  1.85s/it]
                                                         

 72%|███████▏  | 10590/14725 [5:23:36<2:07:35,  1.85s/it]
 72%|███████▏  | 10591/14725 [5:23:37<2:05:45,  1.83s/it]
 72%|███████▏  | 10592/14725 [5:23:39<2:04:20,  1.81s/it]
 72%|███████▏  | 10593/14725 [5:23:41<2:03:56,  1.80s/it]
 72%|██████�
+0: {'loss': 0.8222, 'grad_norm': 0.7306159364238843, 'learning_rate': 2e-05, 'epoch': 0.72}
+0: ��▏  | 10594/14725 [5:23:43<2:03:50,  1.80s/it]
 72%|███████▏  | 10595/14725 [5:23:45<2:02:36,  1.78s/it]
 72%|███████▏  | 10596/14725 [5:23:46<2:03:33,  1.80s/it]
 72%|███████▏  | 10597/14725 [5:23:48<2:03:39,  1.80s/it]
 72%|███████▏  | 10598/14725 [5:23:50<2:04:25,  1.81s/it]
 72%|███████▏  | 10599/14725 [5:23:52<2:05:36,  1.83s/it]
 72%|███████▏  | 10600/14725 [5:23:54<2:04:44,  1.81s/it]
                                                         

 72%|███████▏  | 10600/14725 [5:23:54<2:04:44,  1.81s/it]
 72%|███████▏  | 10601/14725 [5:23:55<2:04:45,  1.82s/it]
 72%|███████▏  | 10602/14725 [5:23:57<2:03:51,  1.80s/it]
 72%|███████▏  | 10603/14725 [5:23:59<2:02:54,  1.79s/it]
 72%|███████▏  | 10604/14725 [5:24:01<2:04:42,  1.82s/it]
 72%|███████▏  | 10605/14725 [5:24:03<2:03:25,  1.80s/it]
 72%|███████�
+0: {'loss': 0.7963, 'grad_norm': 0.6685715189994224, 'learning_rate': 2e-05, 'epoch': 0.72}
+0: ��  | 10606/14725 [5:24:04<2:02:54,  1.79s/it]
 72%|███████▏  | 10607/14725 [5:24:06<2:03:38,  1.80s/it]
 72%|███████▏  | 10608/14725 [5:24:08<2:04:07,  1.81s/it]
 72%|███████▏  | 10609/14725 [5:24:10<2:03:34,  1.80s/it]
 72%|███████▏  | 10610/14725 [5:24:12<2:03:28,  1.80s/it]
                                                         

 72%|███████▏  | 10610/14725 [5:24:12<2:03:28,  1.80s/it]
 72%|███████▏  | 10611/14725 [5:24:13<2:04:12,  1.81s/it]
 72%|███████▏  | 10612/14725 [5:24:15<2:04:22,  1.81s/it]
 72%|███████▏  | 10613/14725 [5:24:17<2:03:41,  1.80s/it]
 72%|███████▏  | 10614/14725 [5:24:19<2:03:58,  1.81s/it]
 72%|███████▏  | 10615/14725 [5:24:21<2:04:40,  1.82s/it]
 72%|███████▏  | 10616/14725 [5:24:23<2:03:11,  1.80s/it]
 72%|███████▏  | 10617/14725 [5:24:24<2:02:56,  1.80s/it]
 72%|███████▏ 
+0: {'loss': 0.7946, 'grad_norm': 0.6806601836294729, 'learning_rate': 2e-05, 'epoch': 0.72}
+0:  | 10618/14725 [5:24:26<2:02:57,  1.80s/it]
 72%|███████▏  | 10619/14725 [5:24:28<2:02:39,  1.79s/it]
 72%|███████▏  | 10620/14725 [5:24:30<2:03:07,  1.80s/it]
                                                         

 72%|███████▏  | 10620/14725 [5:24:30<2:03:07,  1.80s/it]
 72%|███████▏  | 10621/14725 [5:24:31<2:02:05,  1.79s/it]
 72%|███████▏  | 10622/14725 [5:24:33<2:02:17,  1.79s/it]
 72%|███████▏  | 10623/14725 [5:24:35<2:01:31,  1.78s/it]
 72%|███████▏  | 10624/14725 [5:24:37<2:01:36,  1.78s/it]
 72%|███████▏  | 10625/14725 [5:24:39<2:01:31,  1.78s/it]
 72%|███████▏  | 10626/14725 [5:24:40<2:00:55,  1.77s/it]
 72%|███████▏  | 10627/14725 [5:24:42<2:00:46,  1.77s/it]
 72%|███████▏  | 10628/14725 [5:24:44<2:01:35,  1.78s/it]
 72%|███████▏  | 10629/14725 [5:24:46<2:02:17,  1.79s/it]
 72%|███████▏  | 
+0: {'loss': 0.7841, 'grad_norm': 0.7023132321980048, 'learning_rate': 2e-05, 'epoch': 0.72}
+0: {'loss': 0.8036, 'grad_norm': 0.6892943893089707, 'learning_rate': 2e-05, 'epoch': 0.72}
+0: 10630/14725 [5:24:48<2:03:24,  1.81s/it]
                                                         

 72%|███████▏  | 10630/14725 [5:24:48<2:03:24,  1.81s/it]
 72%|███████▏  | 10631/14725 [5:24:49<2:02:40,  1.80s/it]
 72%|███████▏  | 10632/14725 [5:24:51<2:04:54,  1.83s/it]
 72%|███████▏  | 10633/14725 [5:24:53<2:05:21,  1.84s/it]
 72%|███████▏  | 10634/14725 [5:24:55<2:03:44,  1.81s/it]
 72%|███████▏  | 10635/14725 [5:24:57<2:03:06,  1.81s/it]
 72%|███████▏  | 10636/14725 [5:24:58<2:03:30,  1.81s/it]
 72%|███████▏  | 10637/14725 [5:25:00<2:02:15,  1.79s/it]
 72%|███████▏  | 10638/14725 [5:25:02<2:02:13,  1.79s/it]
 72%|███████▏  | 10639/14725 [5:25:04<2:02:41,  1.80s/it]
 72%|███████▏  | 10640/14725 [5:25:06<2:02:53,  1.81s/it]
                                                         

 72%|██████��▏  | 10640/14725 [5:25:
+0: {'loss': 0.8301, 'grad_norm': 0.6694746555323026, 'learning_rate': 2e-05, 'epoch': 0.72}
+0: 06<2:02:53,  1.81s/it]
 72%|███████▏  | 10641/14725 [5:25:08<2:04:19,  1.83s/it]
 72%|███████▏  | 10642/14725 [5:25:09<2:02:33,  1.80s/it]
 72%|███████▏  | 10643/14725 [5:25:11<2:02:20,  1.80s/it]
 72%|███████▏  | 10644/14725 [5:25:13<2:02:47,  1.81s/it]
 72%|███████▏  | 10645/14725 [5:25:15<2:02:53,  1.81s/it]
 72%|███████▏  | 10646/14725 [5:25:17<2:04:23,  1.83s/it]
 72%|███████▏  | 10647/14725 [5:25:18<2:03:14,  1.81s/it]
 72%|███████▏  | 10648/14725 [5:25:20<2:03:12,  1.81s/it]
 72%|███████▏  | 10649/14725 [5:25:22<2:04:44,  1.84s/it]
 72%|███████▏  | 10650/14725 [5:25:24<2:04:10,  1.83s/it]
                                                         

 72%|███████▏  | 10650/14725 [5:25:24<2:04:10,  1.83s/it]
 72%|███████▏  | 10651/14725 [5:25:26<2:03:59,  1.83s/it]
 72%|███████▏  | 10652/14725 [5:25:27<
+0: {'loss': 0.7975, 'grad_norm': 0.6334299553371144, 'learning_rate': 2e-05, 'epoch': 0.72}
+0: 2:02:43,  1.81s/it]
 72%|███████▏  | 10653/14725 [5:25:29<2:03:20,  1.82s/it]
 72%|███████▏  | 10654/14725 [5:25:31<2:03:19,  1.82s/it]
 72%|███████▏  | 10655/14725 [5:25:33<2:03:23,  1.82s/it]
 72%|███████▏  | 10656/14725 [5:25:35<2:02:05,  1.80s/it]
 72%|███████▏  | 10657/14725 [5:25:36<2:01:22,  1.79s/it]
 72%|███████▏  | 10658/14725 [5:25:38<2:01:14,  1.79s/it]
 72%|███████▏  | 10659/14725 [5:25:40<2:00:57,  1.78s/it]
 72%|███████▏  | 10660/14725 [5:25:42<2:01:01,  1.79s/it]
                                                         

 72%|███████▏  | 10660/14725 [5:25:42<2:01:01,  1.79s/it]
 72%|███████▏  | 10661/14725 [5:25:44<2:02:24,  1.81s/it]
 72%|███████▏  | 10662/14725 [5:25:45<2:01:55,  1.80s/it]
 72%|███████▏  | 10663/14725 [5:25:47<2:01:38,  1.80s/it]
 72%|███████▏  | 10664/14725 [5:25:49<2:0
+0: {'loss': 0.7839, 'grad_norm': 0.7259626608127583, 'learning_rate': 2e-05, 'epoch': 0.72}
+0: 2:47,  1.81s/it]
 72%|███████▏  | 10665/14725 [5:25:51<2:02:06,  1.80s/it]
 72%|███████▏  | 10666/14725 [5:25:53<2:01:11,  1.79s/it]
 72%|███████▏  | 10667/14725 [5:25:54<2:02:35,  1.81s/it]
 72%|███████▏  | 10668/14725 [5:25:56<2:03:45,  1.83s/it]
 72%|███████▏  | 10669/14725 [5:25:58<2:04:48,  1.85s/it]
 72%|███████▏  | 10670/14725 [5:26:00<2:05:03,  1.85s/it]
                                                         

 72%|███████▏  | 10670/14725 [5:26:00<2:05:03,  1.85s/it]
 72%|███████▏  | 10671/14725 [5:26:02<2:04:45,  1.85s/it]
 72%|███████▏  | 10672/14725 [5:26:04<2:07:08,  1.88s/it]
 72%|███████▏  | 10673/14725 [5:26:06<2:05:38,  1.86s/it]
 72%|███████▏  | 10674/14725 [5:26:07<2:04:04,  1.84s/it]
 72%|███████▏  | 10675/14725 [5:26:09<2:04:40,  1.85s/it]
 73%|███████▎  | 10676/14725 [5:26:11<2:04:4
+0: {'loss': 0.8094, 'grad_norm': 0.663989693485517, 'learning_rate': 2e-05, 'epoch': 0.73}
+0: 2,  1.85s/it]
 73%|███████▎  | 10677/14725 [5:26:13<2:03:15,  1.83s/it]
 73%|███████▎  | 10678/14725 [5:26:15<2:03:31,  1.83s/it]
 73%|███████▎  | 10679/14725 [5:26:17<2:02:21,  1.81s/it]
 73%|███████▎  | 10680/14725 [5:26:18<2:01:37,  1.80s/it]
                                                         

 73%|███████▎  | 10680/14725 [5:26:18<2:01:37,  1.80s/it]
 73%|███████▎  | 10681/14725 [5:26:20<2:01:35,  1.80s/it]
 73%|███████▎  | 10682/14725 [5:26:22<2:01:26,  1.80s/it]
 73%|███████▎  | 10683/14725 [5:26:24<2:01:26,  1.80s/it]
 73%|███████▎  | 10684/14725 [5:26:26<2:01:02,  1.80s/it]
 73%|███████▎  | 10685/14725 [5:26:27<2:00:58,  1.80s/it]
 73%|███████▎  | 10686/14725 [5:26:29<2:00:06,  1.78s/it]
 73%|███████▎  | 10687/14725 [5:26:31<1:59:51,  1.78s/it]
 73%|███████▎  | 10688/14725 [5:26:33<2:00:55, 
+0: {'loss': 0.799, 'grad_norm': 0.6562156448434502, 'learning_rate': 2e-05, 'epoch': 0.73}
+0:  1.80s/it]
 73%|███████▎  | 10689/14725 [5:26:35<2:00:48,  1.80s/it]
 73%|███████▎  | 10690/14725 [5:26:37<2:04:57,  1.86s/it]
                                                         

 73%|███████▎  | 10690/14725 [5:26:37<2:04:57,  1.86s/it]
 73%|███████▎  | 10691/14725 [5:26:39<2:09:27,  1.93s/it]
 73%|███████▎  | 10692/14725 [5:26:40<2:06:01,  1.87s/it]
 73%|███████▎  | 10693/14725 [5:26:42<2:04:24,  1.85s/it]
 73%|███████▎  | 10694/14725 [5:26:44<2:02:25,  1.82s/it]
 73%|███████▎  | 10695/14725 [5:26:46<2:01:43,  1.81s/it]
 73%|███████▎  | 10696/14725 [5:26:48<2:01:37,  1.81s/it]
 73%|███████▎  | 10697/14725 [5:26:49<2:01:37,  1.81s/it]
 73%|███████▎  | 10698/14725 [5:26:51<2:00:53,  1.80s/it]
 73%|███████▎  | 10699/14725 [5:26:53<2:01:55,  1.82s/it]
 73%|███████▎  | 10700/14725 [5:26:55<2:02:59,  1.
+0: {'loss': 0.801, 'grad_norm': 0.6523619990348358, 'learning_rate': 2e-05, 'epoch': 0.73}
+0: {'loss': 0.8056, 'grad_norm': 0.6573083507871053, 'learning_rate': 2e-05, 'epoch': 0.73}
+0: 83s/it]
                                                         

 73%|███████▎  | 10700/14725 [5:26:55<2:02:59,  1.83s/it]
 73%|███████▎  | 10701/14725 [5:26:57<2:01:57,  1.82s/it]
 73%|███████▎  | 10702/14725 [5:26:59<2:03:27,  1.84s/it]
 73%|███████▎  | 10703/14725 [5:27:00<2:01:57,  1.82s/it]
 73%|███████▎  | 10704/14725 [5:27:02<2:04:01,  1.85s/it]
 73%|███████▎  | 10705/14725 [5:27:04<2:02:28,  1.83s/it]
 73%|███████▎  | 10706/14725 [5:27:06<2:01:38,  1.82s/it]
 73%|███████▎  | 10707/14725 [5:27:08<2:00:24,  1.80s/it]
 73%|███████▎  | 10708/14725 [5:27:09<1:59:40,  1.79s/it]
 73%|███████▎  | 10709/14725 [5:27:11<1:59:10,  1.78s/it]
 73%|███████▎  | 10710/14725 [5:27:13<1:59:43,  1.79s/it]
                                                         

 73%|███████▎  | 10710/14725 [5:27:13<1:59:43,  1.79s/it]
 73%|█�
+0: {'loss': 0.8093, 'grad_norm': 0.6880460152766944, 'learning_rate': 2e-05, 'epoch': 0.73}
+0: �█████▎  | 10711/14725 [5:27:15<2:01:29,  1.82s/it]
 73%|███████▎  | 10712/14725 [5:27:17<2:00:58,  1.81s/it]
 73%|███████▎  | 10713/14725 [5:27:18<2:00:07,  1.80s/it]
 73%|███████▎  | 10714/14725 [5:27:20<2:03:05,  1.84s/it]
 73%|███████▎  | 10715/14725 [5:27:22<2:03:18,  1.84s/it]
 73%|███████▎  | 10716/14725 [5:27:24<2:04:14,  1.86s/it]
 73%|███████▎  | 10717/14725 [5:27:26<2:02:14,  1.83s/it]
 73%|███████▎  | 10718/14725 [5:27:28<2:01:57,  1.83s/it]
 73%|███████▎  | 10719/14725 [5:27:29<2:00:30,  1.80s/it]
 73%|███████▎  | 10720/14725 [5:27:31<2:01:58,  1.83s/it]
                                                         

 73%|███████▎  | 10720/14725 [5:27:31<2:01:58,  1.83s/it]
 73%|███████▎  | 10721/14725 [5:27:33<2:01:45,  1.82s/it]
 73%|███████▎  | 10722/14725 [5:27:35<2:01:48,  1.83s/it]
 73%|██�
+0: {'loss': 0.7983, 'grad_norm': 0.718650771352783, 'learning_rate': 2e-05, 'epoch': 0.73}
+0: �████▎  | 10723/14725 [5:27:37<2:00:47,  1.81s/it]
 73%|███████▎  | 10724/14725 [5:27:39<2:03:42,  1.86s/it]
 73%|███████▎  | 10725/14725 [5:27:40<2:02:13,  1.83s/it]
 73%|███████▎  | 10726/14725 [5:27:42<2:01:31,  1.82s/it]
 73%|███████▎  | 10727/14725 [5:27:44<2:00:14,  1.80s/it]
 73%|███████▎  | 10728/14725 [5:27:46<2:01:43,  1.83s/it]
 73%|███████▎  | 10729/14725 [5:27:48<2:02:24,  1.84s/it]
 73%|███████▎  | 10730/14725 [5:27:50<2:04:04,  1.86s/it]
                                                         

 73%|███████▎  | 10730/14725 [5:27:50<2:04:04,  1.86s/it]
 73%|███████▎  | 10731/14725 [5:27:51<2:02:24,  1.84s/it]
 73%|███████▎  | 10732/14725 [5:27:53<2:02:43,  1.84s/it]
 73%|███████▎  | 10733/14725 [5:27:55<2:01:04,  1.82s/it]
 73%|███████▎  | 10734/14725 [5:27:57<2:03:06,  1.85s/it]
 73%|███�
+0: {'loss': 0.7827, 'grad_norm': 0.7159177591308615, 'learning_rate': 2e-05, 'epoch': 0.73}
+0: �███▎  | 10735/14725 [5:27:59<2:00:57,  1.82s/it]
 73%|███████▎  | 10736/14725 [5:28:00<1:59:52,  1.80s/it]
 73%|███████▎  | 10737/14725 [5:28:02<2:00:42,  1.82s/it]
 73%|███████▎  | 10738/14725 [5:28:04<2:00:32,  1.81s/it]
 73%|███████▎  | 10739/14725 [5:28:06<2:01:29,  1.83s/it]
 73%|███████▎  | 10740/14725 [5:28:08<2:01:29,  1.83s/it]
                                                         

 73%|███████▎  | 10740/14725 [5:28:08<2:01:29,  1.83s/it]
 73%|███████▎  | 10741/14725 [5:28:10<2:00:54,  1.82s/it]
 73%|███████▎  | 10742/14725 [5:28:11<2:00:03,  1.81s/it]
 73%|███████▎  | 10743/14725 [5:28:13<1:59:53,  1.81s/it]
 73%|███████▎  | 10744/14725 [5:28:15<2:03:11,  1.86s/it]
 73%|███████▎  | 10745/14725 [5:28:17<2:01:43,  1.83s/it]
 73%|███████▎  | 10746/14725 [5:28:19<2:01:26,  1.83s/it]
 73%|████�
+0: {'loss': 0.7958, 'grad_norm': 0.6805399162644905, 'learning_rate': 2e-05, 'epoch': 0.73}
+0: �██▎  | 10747/14725 [5:28:21<2:02:28,  1.85s/it]
 73%|███████▎  | 10748/14725 [5:28:22<2:01:44,  1.84s/it]
 73%|███████▎  | 10749/14725 [5:28:24<2:01:24,  1.83s/it]
 73%|███████▎  | 10750/14725 [5:28:26<2:01:06,  1.83s/it]
                                                         

 73%|███████▎  | 10750/14725 [5:28:26<2:01:06,  1.83s/it]
 73%|███████▎  | 10751/14725 [5:28:28<2:00:23,  1.82s/it]
 73%|███████▎  | 10752/14725 [5:28:30<1:59:29,  1.80s/it]
 73%|███████▎  | 10753/14725 [5:28:31<1:58:07,  1.78s/it]
 73%|███████▎  | 10754/14725 [5:28:33<1:59:15,  1.80s/it]
 73%|███████▎  | 10755/14725 [5:28:35<1:59:05,  1.80s/it]
 73%|███████▎  | 10756/14725 [5:28:37<2:03:51,  1.87s/it]
 73%|███████▎  | 10757/14725 [5:28:39<2:04:27,  1.88s/it]
 73%|███████▎  | 10758/14725 [5:28:41<2:03:05,  1.86s/it]
 73%|█████�
+0: {'loss': 0.8034, 'grad_norm': 0.6865382205878777, 'learning_rate': 2e-05, 'epoch': 0.73}
+0: �█▎  | 10759/14725 [5:28:43<2:01:27,  1.84s/it]
 73%|███████▎  | 10760/14725 [5:28:44<2:01:26,  1.84s/it]
                                                         

 73%|███████▎  | 10760/14725 [5:28:44<2:01:26,  1.84s/it]
 73%|███████▎  | 10761/14725 [5:28:46<2:01:03,  1.83s/it]
 73%|███████▎  | 10762/14725 [5:28:48<2:00:05,  1.82s/it]
 73%|███████▎  | 10763/14725 [5:28:50<2:00:56,  1.83s/it]
 73%|███████▎  | 10764/14725 [5:28:52<2:02:16,  1.85s/it]
 73%|███████▎  | 10765/14725 [5:28:54<2:01:12,  1.84s/it]
 73%|███████▎  | 10766/14725 [5:28:55<2:00:10,  1.82s/it]
 73%|███████▎  | 10767/14725 [5:28:57<1:59:52,  1.82s/it]
 73%|███████▎  | 10768/14725 [5:28:59<2:00:27,  1.83s/it]
 73%|███████▎  | 10769/14725 [5:29:01<1:59:40,  1.82s/it]
 73%|███████▎  | 10770/14725 [5:29:03<2:01:03,  1.84s/it]
                         
+0: {'loss': 0.8207, 'grad_norm': 0.6776930406903791, 'learning_rate': 2e-05, 'epoch': 0.73}
+0: {'loss': 0.7905, 'grad_norm': 0.6971687462841244, 'learning_rate': 2e-05, 'epoch': 0.73}
+0:                                 

 73%|███████▎  | 10770/14725 [5:29:03<2:01:03,  1.84s/it]
 73%|███████▎  | 10771/14725 [5:29:04<2:00:23,  1.83s/it]
 73%|███████▎  | 10772/14725 [5:29:06<1:58:42,  1.80s/it]
 73%|███████▎  | 10773/14725 [5:29:08<1:59:17,  1.81s/it]
 73%|███████▎  | 10774/14725 [5:29:10<1:58:14,  1.80s/it]
 73%|███████▎  | 10775/14725 [5:29:12<1:58:00,  1.79s/it]
 73%|███████▎  | 10776/14725 [5:29:13<1:57:09,  1.78s/it]
 73%|███████▎  | 10777/14725 [5:29:15<1:58:32,  1.80s/it]
 73%|███████▎  | 10778/14725 [5:29:17<1:59:14,  1.81s/it]
 73%|███████▎  | 10779/14725 [5:29:19<2:00:16,  1.83s/it]
 73%|███████▎  | 10780/14725 [5:29:21<2:00:01,  1.83s/it]
                                                         

 73%|███████▎  | 10780/14725 [5:29:21<2:00:01,  1.83s/it]
 73%|███████▎  | 10781/1472
+0: {'loss': 0.8246, 'grad_norm': 0.6667523505140973, 'learning_rate': 2e-05, 'epoch': 0.73}
+0: 5 [5:29:23<1:59:44,  1.82s/it]
 73%|███████▎  | 10782/14725 [5:29:24<1:59:50,  1.82s/it]
 73%|███████▎  | 10783/14725 [5:29:26<1:58:40,  1.81s/it]
 73%|███████▎  | 10784/14725 [5:29:28<1:59:07,  1.81s/it]
 73%|███████▎  | 10785/14725 [5:29:30<1:58:29,  1.80s/it]
 73%|███████▎  | 10786/14725 [5:29:32<1:58:39,  1.81s/it]
 73%|███████▎  | 10787/14725 [5:29:33<2:00:41,  1.84s/it]
 73%|███████▎  | 10788/14725 [5:29:35<1:58:51,  1.81s/it]
 73%|███████▎  | 10789/14725 [5:29:37<1:59:38,  1.82s/it]
 73%|███████▎  | 10790/14725 [5:29:39<1:59:56,  1.83s/it]
                                                         

 73%|███████▎  | 10790/14725 [5:29:39<1:59:56,  1.83s/it]
 73%|███████▎  | 10791/14725 [5:29:41<1:59:06,  1.82s/it]
 73%|███████▎  | 10792/14725 [5:29:42<1:57:49,  1.80s/it]
 73%|███████▎  | 10793/14725 [
+0: {'loss': 0.8134, 'grad_norm': 0.6806858672664492, 'learning_rate': 2e-05, 'epoch': 0.73}
+0: 5:29:44<1:59:11,  1.82s/it]
 73%|███████▎  | 10794/14725 [5:29:46<1:58:43,  1.81s/it]
 73%|███████▎  | 10795/14725 [5:29:48<1:59:50,  1.83s/it]
 73%|███████▎  | 10796/14725 [5:29:50<1:59:59,  1.83s/it]
 73%|███████▎  | 10797/14725 [5:29:52<1:58:45,  1.81s/it]
 73%|███████▎  | 10798/14725 [5:29:53<1:59:09,  1.82s/it]
 73%|███████▎  | 10799/14725 [5:29:55<1:58:25,  1.81s/it]
 73%|███████▎  | 10800/14725 [5:29:57<1:57:33,  1.80s/it]
                                                         

 73%|███████▎  | 10800/14725 [5:29:57<1:57:33,  1.80s/it]
 73%|███████▎  | 10801/14725 [5:29:59<1:57:30,  1.80s/it]
 73%|███████▎  | 10802/14725 [5:30:01<1:58:35,  1.81s/it]
 73%|███████▎  | 10803/14725 [5:30:02<1:57:08,  1.79s/it]
 73%|███████▎  | 10804/14725 [5:30:04<1:57:14,  1.79s/it]
 73%|███████▎  | 10805/14725 [5:3
+0: {'loss': 0.8097, 'grad_norm': 0.6893845377364011, 'learning_rate': 2e-05, 'epoch': 0.73}
+0: 0:06<1:57:27,  1.80s/it]
 73%|███████▎  | 10806/14725 [5:30:08<1:57:24,  1.80s/it]
 73%|███████▎  | 10807/14725 [5:30:10<1:58:16,  1.81s/it]
 73%|███████▎  | 10808/14725 [5:30:11<1:57:24,  1.80s/it]
 73%|███████▎  | 10809/14725 [5:30:13<1:59:26,  1.83s/it]
 73%|███████▎  | 10810/14725 [5:30:15<1:59:43,  1.83s/it]
                                                         

 73%|███████▎  | 10810/14725 [5:30:15<1:59:43,  1.83s/it]
 73%|███████▎  | 10811/14725 [5:30:17<1:59:31,  1.83s/it]
 73%|███████▎  | 10812/14725 [5:30:19<1:57:53,  1.81s/it]
 73%|███████▎  | 10813/14725 [5:30:21<1:58:15,  1.81s/it]
 73%|███████▎  | 10814/14725 [5:30:22<1:57:18,  1.80s/it]
 73%|███████▎  | 10815/14725 [5:30:24<1:57:48,  1.81s/it]
 73%|███████▎  | 10816/14725 [5:30:26<1:59:04,  1.83s/it]
 73%|███████▎  | 10817/14725 [5:30:2
+0: {'loss': 0.8074, 'grad_norm': 0.6480951423940161, 'learning_rate': 2e-05, 'epoch': 0.73}
+0: 8<1:59:38,  1.84s/it]
 73%|███████▎  | 10818/14725 [5:30:30<1:59:29,  1.84s/it]
 73%|███████▎  | 10819/14725 [5:30:31<1:58:29,  1.82s/it]
 73%|███████▎  | 10820/14725 [5:30:33<1:57:01,  1.80s/it]
                                                         

 73%|███████▎  | 10820/14725 [5:30:33<1:57:01,  1.80s/it]
 73%|███████▎  | 10821/14725 [5:30:35<1:56:59,  1.80s/it]
 73%|███████▎  | 10822/14725 [5:30:37<1:56:41,  1.79s/it]
 74%|███████▎  | 10823/14725 [5:30:39<1:56:53,  1.80s/it]
 74%|███████▎  | 10824/14725 [5:30:40<1:56:58,  1.80s/it]
 74%|███████▎  | 10825/14725 [5:30:42<1:57:59,  1.82s/it]
 74%|███████▎  | 10826/14725 [5:30:44<1:57:14,  1.80s/it]
 74%|███████▎  | 10827/14725 [5:30:46<1:56:10,  1.79s/it]
 74%|███████▎  | 10828/14725 [5:30:48<1:56:06,  1.79s/it]
 74%|███████▎  | 10829/14725 [5:30:49<1
+0: {'loss': 0.8161, 'grad_norm': 0.7246810986835648, 'learning_rate': 2e-05, 'epoch': 0.74}
+0: :57:16,  1.81s/it]
 74%|███████▎  | 10830/14725 [5:30:51<1:57:40,  1.81s/it]
                                                         

 74%|███████▎  | 10830/14725 [5:30:51<1:57:40,  1.81s/it]
 74%|███████▎  | 10831/14725 [5:30:53<1:58:43,  1.83s/it]
 74%|███████▎  | 10832/14725 [5:30:55<1:57:03,  1.80s/it]
 74%|███████▎  | 10833/14725 [5:30:57<1:55:49,  1.79s/it]
 74%|███████▎  | 10834/14725 [5:30:58<1:56:23,  1.79s/it]
 74%|███████▎  | 10835/14725 [5:31:00<1:55:37,  1.78s/it]
 74%|███████▎  | 10836/14725 [5:31:02<1:56:24,  1.80s/it]
 74%|███████▎  | 10837/14725 [5:31:04<1:56:15,  1.79s/it]
 74%|███████▎  | 10838/14725 [5:31:06<1:56:14,  1.79s/it]
 74%|███████▎  | 10839/14725 [5:31:07<1:56:49,  1.80s/it]
 74%|███████▎  | 10840/14725 [5:31:09<1:57:01,  1.81s/it]
                                                         
+0: {'loss': 0.8043, 'grad_norm': 0.6826945853700623, 'learning_rate': 2e-05, 'epoch': 0.74}
+0: {'loss': 0.781, 'grad_norm': 0.6718366753708541, 'learning_rate': 2e-05, 'epoch': 0.74}
+0: 
 74%|███████▎  | 10840/14725 [5:31:09<1:57:01,  1.81s/it]
 74%|███████▎  | 10841/14725 [5:31:11<1:56:20,  1.80s/it]
 74%|███████▎  | 10842/14725 [5:31:13<1:56:13,  1.80s/it]
 74%|███████▎  | 10843/14725 [5:31:15<1:55:42,  1.79s/it]
 74%|███████▎  | 10844/14725 [5:31:16<1:54:59,  1.78s/it]
 74%|███████▎  | 10845/14725 [5:31:18<1:56:14,  1.80s/it]
 74%|███████▎  | 10846/14725 [5:31:20<1:56:37,  1.80s/it]
 74%|███████▎  | 10847/14725 [5:31:22<1:56:08,  1.80s/it]
 74%|███████▎  | 10848/14725 [5:31:24<1:56:51,  1.81s/it]
 74%|███████▎  | 10849/14725 [5:31:25<1:56:13,  1.80s/it]
 74%|███████▎  | 10850/14725 [5:31:27<1:56:58,  1.81s/it]
                                                         

 74%|███████▎  | 10850/14725 [5:31:27<1:56:58,  1.81s/it]
 74%|███████▎  | 10851/14725 [5:31:29<1:57:01,  1.81s/it]
 7
+0: {'loss': 0.8164, 'grad_norm': 0.6825784403809828, 'learning_rate': 2e-05, 'epoch': 0.74}
+0: 4%|███████▎  | 10852/14725 [5:31:31<1:57:07,  1.81s/it]
 74%|███████▎  | 10853/14725 [5:31:33<1:55:47,  1.79s/it]
 74%|███████▎  | 10854/14725 [5:31:34<1:55:45,  1.79s/it]
 74%|███████▎  | 10855/14725 [5:31:36<1:56:11,  1.80s/it]
 74%|███████▎  | 10856/14725 [5:31:38<1:56:02,  1.80s/it]
 74%|███████▎  | 10857/14725 [5:31:40<1:56:23,  1.81s/it]
 74%|███████▎  | 10858/14725 [5:31:42<1:56:26,  1.81s/it]
 74%|███████▎  | 10859/14725 [5:31:43<1:56:11,  1.80s/it]
 74%|███████▍  | 10860/14725 [5:31:45<1:55:52,  1.80s/it]
                                                         

 74%|███████▍  | 10860/14725 [5:31:45<1:55:52,  1.80s/it]
 74%|███████▍  | 10861/14725 [5:31:47<1:57:02,  1.82s/it]
 74%|███████▍  | 10862/14725 [5:31:49<1:58:08,  1.83s/it]
 74%|███████▍  | 10863/14725 [5:31:51<1:57:46,  1.83s/it]
 74%|
+0: {'loss': 0.8088, 'grad_norm': 0.6547629353655156, 'learning_rate': 2e-05, 'epoch': 0.74}
+0: ███████▍  | 10864/14725 [5:31:53<1:59:10,  1.85s/it]
 74%|███████▍  | 10865/14725 [5:31:55<1:58:51,  1.85s/it]
 74%|███████▍  | 10866/14725 [5:31:56<1:57:45,  1.83s/it]
 74%|███████▍  | 10867/14725 [5:31:58<1:57:47,  1.83s/it]
 74%|███████▍  | 10868/14725 [5:32:00<1:56:13,  1.81s/it]
 74%|███████▍  | 10869/14725 [5:32:02<1:55:09,  1.79s/it]
 74%|███████▍  | 10870/14725 [5:32:03<1:54:31,  1.78s/it]
                                                         

 74%|███████▍  | 10870/14725 [5:32:03<1:54:31,  1.78s/it]
 74%|███████▍  | 10871/14725 [5:32:05<1:55:04,  1.79s/it]
 74%|███████▍  | 10872/14725 [5:32:07<1:56:26,  1.81s/it]
 74%|███████▍  | 10873/14725 [5:32:09<1:57:16,  1.83s/it]
 74%|███████▍  | 10874/14725 [5:32:11<1:57:59,  1.84s/it]
 74%|███████▍  | 10875/14725 [5:32:13<1:56:50,  1.82s/it]
 74%|█
+0: {'loss': 0.7922, 'grad_norm': 0.7255873438098311, 'learning_rate': 2e-05, 'epoch': 0.74}
+0: ██████▍  | 10876/14725 [5:32:14<1:57:32,  1.83s/it]
 74%|███████▍  | 10877/14725 [5:32:16<1:58:28,  1.85s/it]
 74%|███████▍  | 10878/14725 [5:32:18<1:58:21,  1.85s/it]
 74%|███████▍  | 10879/14725 [5:32:20<1:58:12,  1.84s/it]
 74%|███████▍  | 10880/14725 [5:32:22<1:58:53,  1.86s/it]
                                                         

 74%|███████▍  | 10880/14725 [5:32:22<1:58:53,  1.86s/it]
 74%|███████▍  | 10881/14725 [5:32:24<1:57:46,  1.84s/it]
 74%|███████▍  | 10882/14725 [5:32:26<1:57:19,  1.83s/it]
 74%|███████▍  | 10883/14725 [5:32:27<1:56:12,  1.81s/it]
 74%|███████▍  | 10884/14725 [5:32:29<1:55:09,  1.80s/it]
 74%|███████▍  | 10885/14725 [5:32:31<1:55:09,  1.80s/it]
 74%|███████▍  | 10886/14725 [5:32:33<1:54:59,  1.80s/it]
 74%|███████▍  | 10887/14725 [5:32:34<1:53:59,  1.78s/it]
 74%|██
+0: {'loss': 0.7992, 'grad_norm': 0.6725768300749075, 'learning_rate': 2e-05, 'epoch': 0.74}
+0: █████▍  | 10888/14725 [5:32:36<1:53:04,  1.77s/it]
 74%|███████▍  | 10889/14725 [5:32:38<1:53:15,  1.77s/it]
 74%|███████▍  | 10890/14725 [5:32:40<1:54:41,  1.79s/it]
                                                         

 74%|███████▍  | 10890/14725 [5:32:40<1:54:41,  1.79s/it]
 74%|███████▍  | 10891/14725 [5:32:42<1:57:28,  1.84s/it]
 74%|███████▍  | 10892/14725 [5:32:44<1:57:21,  1.84s/it]
 74%|███████▍  | 10893/14725 [5:32:45<1:58:20,  1.85s/it]
 74%|███████▍  | 10894/14725 [5:32:47<1:57:40,  1.84s/it]
 74%|███████▍  | 10895/14725 [5:32:49<1:56:23,  1.82s/it]
 74%|███████▍  | 10896/14725 [5:32:51<1:56:00,  1.82s/it]
 74%|███████▍  | 10897/14725 [5:32:53<1:55:31,  1.81s/it]
 74%|███████▍  | 10898/14725 [5:32:54<1:55:48,  1.82s/it]
 74%|███████▍  | 10899/14725 [5:32:56<1:55:55,  1.82s/it]
 74%|███
+0: {'loss': 0.7944, 'grad_norm': 0.7301793458373623, 'learning_rate': 2e-05, 'epoch': 0.74}
+0: {'loss': 0.7998, 'grad_norm': 0.7054134846181959, 'learning_rate': 2e-05, 'epoch': 0.74}
+0: ████▍  | 10900/14725 [5:32:58<1:56:16,  1.82s/it]
                                                         

 74%|███████▍  | 10900/14725 [5:32:58<1:56:16,  1.82s/it]
 74%|███████▍  | 10901/14725 [5:33:00<1:55:23,  1.81s/it]
 74%|███████▍  | 10902/14725 [5:33:02<1:54:55,  1.80s/it]
 74%|███████▍  | 10903/14725 [5:33:03<1:54:14,  1.79s/it]
 74%|█��█████▍  | 10904/14725 [5:33:05<1:54:05,  1.79s/it]
 74%|███████▍  | 10905/14725 [5:33:07<1:54:21,  1.80s/it]
 74%|███████▍  | 10906/14725 [5:33:09<1:55:55,  1.82s/it]
 74%|███████▍  | 10907/14725 [5:33:11<1:54:34,  1.80s/it]
 74%|███████▍  | 10908/14725 [5:33:12<1:53:42,  1.79s/it]
 74%|███████▍  | 10909/14725 [5:33:14<1:53:54,  1.79s/it]
 74%|███████▍  | 10910/14725 [5:33:16<1:53:30,  1.79s/it]
                                                         

 74%|███████▍  |
+0: {'loss': 0.7906, 'grad_norm': 0.7481891403659623, 'learning_rate': 2e-05, 'epoch': 0.74}
+0:  10910/14725 [5:33:16<1:53:30,  1.79s/it]
 74%|███████▍  | 10911/14725 [5:33:18<1:53:25,  1.78s/it]
 74%|███████▍  | 10912/14725 [5:33:20<1:53:33,  1.79s/it]
 74%|███████▍  | 10913/14725 [5:33:21<1:52:35,  1.77s/it]
 74%|███████▍  | 10914/14725 [5:33:23<1:53:49,  1.79s/it]
 74%|███████▍  | 10915/14725 [5:33:25<1:52:58,  1.78s/it]
 74%|███████▍  | 10916/14725 [5:33:27<1:54:10,  1.80s/it]
 74%|███████▍  | 10917/14725 [5:33:29<1:55:14,  1.82s/it]
 74%|███████▍  | 10918/14725 [5:33:30<1:54:34,  1.81s/it]
 74%|███████▍  | 10919/14725 [5:33:32<1:53:12,  1.78s/it]
 74%|███████▍  | 10920/14725 [5:33:34<1:53:25,  1.79s/it]
                                                         

 74%|███████▍  | 10920/14725 [5:33:34<1:53:25,  1.79s/it]
 74%|███████▍  | 10921/14725 [5:33:36<1:55:04,  1.82s/it]
 74%|███████▍  | 10
+0: {'loss': 0.796, 'grad_norm': 0.7478403706111587, 'learning_rate': 2e-05, 'epoch': 0.74}
+0: 922/14725 [5:33:38<1:55:11,  1.82s/it]
 74%|███████▍  | 10923/14725 [5:33:40<1:56:53,  1.84s/it]
 74%|███████▍  | 10924/14725 [5:33:41<1:58:10,  1.87s/it]
 74%|███████▍  | 10925/14725 [5:33:43<1:57:01,  1.85s/it]
 74%|███████▍  | 10926/14725 [5:33:45<1:54:52,  1.81s/it]
 74%|███████▍  | 10927/14725 [5:33:47<1:53:41,  1.80s/it]
 74%|███████▍  | 10928/14725 [5:33:49<1:52:50,  1.78s/it]
 74%|███████▍  | 10929/14725 [5:33:50<1:53:24,  1.79s/it]
 74%|███████▍  | 10930/14725 [5:33:52<1:53:12,  1.79s/it]
                                                         

 74%|███████▍  | 10930/14725 [5:33:52<1:53:12,  1.79s/it]
 74%|███████▍  | 10931/14725 [5:33:54<1:54:02,  1.80s/it]
 74%|███████▍  | 10932/14725 [5:33:56<1:55:00,  1.82s/it]
 74%|███████▍  | 10933/14725 [5:33:58<1:54:43,  1.82s/it]
 74%|███████▍  | 10934
+0: {'loss': 0.7731, 'grad_norm': 0.6619785840640139, 'learning_rate': 2e-05, 'epoch': 0.74}
+0: /14725 [5:33:59<1:53:36,  1.80s/it]
 74%|███████▍  | 10935/14725 [5:34:01<1:54:36,  1.81s/it]
 74%|███████▍  | 10936/14725 [5:34:03<1:54:33,  1.81s/it]
 74%|███████▍  | 10937/14725 [5:34:05<1:53:52,  1.80s/it]
 74%|███████▍  | 10938/14725 [5:34:07<1:54:24,  1.81s/it]
 74%|███████▍  | 10939/14725 [5:34:08<1:54:37,  1.82s/it]
 74%|███████▍  | 10940/14725 [5:34:10<1:53:44,  1.80s/it]
                                                         

 74%|███████▍  | 10940/14725 [5:34:10<1:53:44,  1.80s/it]
 74%|███████▍  | 10941/14725 [5:34:12<1:53:22,  1.80s/it]
 74%|███████▍  | 10942/14725 [5:34:14<1:52:51,  1.79s/it]
 74%|███████▍  | 10943/14725 [5:34:16<1:52:35,  1.79s/it]
 74%|███████▍  | 10944/14725 [5:34:17<1:53:36,  1.80s/it]
 74%|███████▍  | 10945/14725 [5:34:19<1:53:11,  1.80s/it]
 74%|███████▍  | 10946/14
+0: {'loss': 0.8061, 'grad_norm': 0.7291958311484732, 'learning_rate': 2e-05, 'epoch': 0.74}
+0: 725 [5:34:21<1:56:05,  1.84s/it]
 74%|███████▍  | 10947/14725 [5:34:23<1:55:47,  1.84s/it]
 74%|███████▍  | 10948/14725 [5:34:25<1:57:24,  1.87s/it]
 74%|███████▍  | 10949/14725 [5:34:27<1:56:23,  1.85s/it]
 74%|███████▍  | 10950/14725 [5:34:29<1:56:07,  1.85s/it]
                                                         

 74%|███████▍  | 10950/14725 [5:34:29<1:56:07,  1.85s/it]
 74%|███████▍  | 10951/14725 [5:34:30<1:56:50,  1.86s/it]
 74%|███████▍  | 10952/14725 [5:34:32<1:55:49,  1.84s/it]
 74%|███████▍  | 10953/14725 [5:34:34<1:54:00,  1.81s/it]
 74%|███████▍  | 10954/14725 [5:34:36<1:53:07,  1.80s/it]
 74%|███████▍  | 10955/14725 [5:34:38<1:54:06,  1.82s/it]
 74%|███████▍  | 10956/14725 [5:34:39<1:53:38,  1.81s/it]
 74%|███████▍  | 10957/14725 [5:34:41<1:54:56,  1.83s/it]
 74%|███████▍  | 10958/14725
+0: {'loss': 0.8132, 'grad_norm': 0.7165374228943802, 'learning_rate': 2e-05, 'epoch': 0.74}
+0:  [5:34:43<1:53:44,  1.81s/it]
 74%|███████▍  | 10959/14725 [5:34:45<1:53:02,  1.80s/it]
 74%|███████▍  | 10960/14725 [5:34:47<1:52:34,  1.79s/it]
                                                         

 74%|███████▍  | 10960/14725 [5:34:47<1:52:34,  1.79s/it]
 74%|███████▍  | 10961/14725 [5:34:48<1:52:20,  1.79s/it]
 74%|███████▍  | 10962/14725 [5:34:50<1:52:33,  1.79s/it]
 74%|███████▍  | 10963/14725 [5:34:52<1:51:56,  1.79s/it]
 74%|███████▍  | 10964/14725 [5:34:54<1:53:03,  1.80s/it]
 74%|███████▍  | 10965/14725 [5:34:56<1:54:08,  1.82s/it]
 74%|███████▍  | 10966/14725 [5:34:58<1:54:58,  1.84s/it]
 74%|███████▍  | 10967/14725 [5:34:59<1:54:15,  1.82s/it]
 74%|███████▍  | 10968/14725 [5:35:01<1:53:50,  1.82s/it]
 74%|███████▍  | 10969/14725 [5:35:03<1:53:06,  1.81s/it]
 74%|███████▍  | 10970/14725 [5
+0: {'loss': 0.799, 'grad_norm': 0.6717859424974263, 'learning_rate': 2e-05, 'epoch': 0.74}
+0: {'loss': 0.7983, 'grad_norm': 0.6841161093435814, 'learning_rate': 2e-05, 'epoch': 0.75}
+0: :35:05<1:54:09,  1.82s/it]
                                                         

 74%|███████▍  | 10970/14725 [5:35:05<1:54:09,  1.82s/it]
 75%|███████▍  | 10971/14725 [5:35:07<1:52:47,  1.80s/it]
 75%|███████▍  | 10972/14725 [5:35:08<1:54:45,  1.83s/it]
 75%|███████▍  | 10973/14725 [5:35:10<1:54:19,  1.83s/it]
 75%|███████▍  | 10974/14725 [5:35:12<1:53:19,  1.81s/it]
 75%|███████▍  | 10975/14725 [5:35:14<1:53:46,  1.82s/it]
 75%|███████▍  | 10976/14725 [5:35:16<1:53:13,  1.81s/it]
 75%|███████▍  | 10977/14725 [5:35:18<1:54:10,  1.83s/it]
 75%|███████▍  | 10978/14725 [5:35:19<1:53:22,  1.82s/it]
 75%|███████▍  | 10979/14725 [5:35:21<1:53:54,  1.82s/it]
 75%|███████▍  | 10980/14725 [5:35:23<1:52:26,  1.80s/it]
                                                         

 75%|███████▍  | 10980/14725 [5:35:23<1:52:26,  1
+0: {'loss': 0.8265, 'grad_norm': 0.6858043673911087, 'learning_rate': 2e-05, 'epoch': 0.75}
+0: .80s/it]
 75%|███████▍  | 10981/14725 [5:35:25<1:52:17,  1.80s/it]
 75%|███████▍  | 10982/14725 [5:35:27<1:52:42,  1.81s/it]
 75%|███████▍  | 10983/14725 [5:35:28<1:53:37,  1.82s/it]
 75%|███████▍  | 10984/14725 [5:35:30<1:52:31,  1.80s/it]
 75%|███████▍  | 10985/14725 [5:35:32<1:53:33,  1.82s/it]
 75%|███████▍  | 10986/14725 [5:35:34<1:52:47,  1.81s/it]
 75%|███████▍  | 10987/14725 [5:35:36<1:54:35,  1.84s/it]
 75%|███████▍  | 10988/14725 [5:35:38<1:54:51,  1.84s/it]
 75%|███████▍  | 10989/14725 [5:35:39<1:52:59,  1.81s/it]
 75%|███████▍  | 10990/14725 [5:35:41<1:51:32,  1.79s/it]
                                                         

 75%|███████▍  | 10990/14725 [5:35:41<1:51:32,  1.79s/it]
 75%|███████▍  | 10991/14725 [5:35:43<1:52:18,  1.80s/it]
 75%|███████▍  | 10992/14725 [5:35:45<1:51:59,  1.80
+0: {'loss': 0.8086, 'grad_norm': 0.6763203511356575, 'learning_rate': 2e-05, 'epoch': 0.75}
+0: s/it]
 75%|███████▍  | 10993/14725 [5:35:46<1:51:20,  1.79s/it]
 75%|███████▍  | 10994/14725 [5:35:48<1:51:31,  1.79s/it]
 75%|███████▍  | 10995/14725 [5:35:50<1:51:19,  1.79s/it]
 75%|███████▍  | 10996/14725 [5:35:52<1:51:33,  1.79s/it]
 75%|███████▍  | 10997/14725 [5:35:54<1:52:26,  1.81s/it]
 75%|███████▍  | 10998/14725 [5:35:56<1:53:04,  1.82s/it]
 75%|███████▍  | 10999/14725 [5:35:57<1:53:06,  1.82s/it]
 75%|███████▍  | 11000/14725 [5:35:59<1:52:22,  1.81s/it]
                                                         

 75%|███████▍  | 11000/14725 [5:35:59<1:52:22,  1.81s/it]
 75%|███████▍  | 11001/14725 [5:36:01<1:51:25,  1.80s/it]
 75%|███████▍  | 11002/14725 [5:36:03<1:51:19,  1.79s/it]
 75%|███████▍  | 11003/14725 [5:36:04<1:51:13,  1.79s/it]
 75%|███████▍  | 11004/14725 [5:36:06<1:50:29,  1.78s/i
+0: {'loss': 0.7735, 'grad_norm': 0.682235540574088, 'learning_rate': 2e-05, 'epoch': 0.75}
+0: t]
 75%|███████▍  | 11005/14725 [5:36:08<1:49:42,  1.77s/it]
 75%|███████▍  | 11006/14725 [5:36:10<1:49:39,  1.77s/it]
 75%|███████▍  | 11007/14725 [5:36:12<1:50:05,  1.78s/it]
 75%|███████▍  | 11008/14725 [5:36:13<1:52:32,  1.82s/it]
 75%|███████▍  | 11009/14725 [5:36:15<1:53:14,  1.83s/it]
 75%|███████▍  | 11010/14725 [5:36:17<1:52:02,  1.81s/it]
                                                         

 75%|███████▍  | 11010/14725 [5:36:17<1:52:02,  1.81s/it]
 75%|███████▍  | 11011/14725 [5:36:19<1:51:46,  1.81s/it]
 75%|███████▍  | 11012/14725 [5:36:21<1:51:47,  1.81s/it]
 75%|███████▍  | 11013/14725 [5:36:22<1:51:07,  1.80s/it]
 75%|███████▍  | 11014/14725 [5:36:24<1:50:18,  1.78s/it]
 75%|███████▍  | 11015/14725 [5:36:26<1:49:40,  1.77s/it]
 75%|███████▍  | 11016/14725 [5:36:28<1:50:12,  1.78s/it]
+0: {'loss': 0.7814, 'grad_norm': 0.6823506733688577, 'learning_rate': 2e-05, 'epoch': 0.75}
+0:  75%|███████▍  | 11017/14725 [5:36:30<1:49:57,  1.78s/it]
 75%|███████▍  | 11018/14725 [5:36:31<1:49:29,  1.77s/it]
 75%|███████▍  | 11019/14725 [5:36:33<1:50:49,  1.79s/it]
 75%|███████▍  | 11020/14725 [5:36:35<1:51:33,  1.81s/it]
                                                         

 75%|███████▍  | 11020/14725 [5:36:35<1:51:33,  1.81s/it]
 75%|███████▍  | 11021/14725 [5:36:37<1:51:33,  1.81s/it]
 75%|███████▍  | 11022/14725 [5:36:39<1:50:38,  1.79s/it]
 75%|███████▍  | 11023/14725 [5:36:40<1:51:23,  1.81s/it]
 75%|███████▍  | 11024/14725 [5:36:42<1:51:59,  1.82s/it]
 75%|███████▍  | 11025/14725 [5:36:44<1:51:29,  1.81s/it]
 75%|███████▍  | 11026/14725 [5:36:46<1:53:02,  1.83s/it]
 75%|███████▍  | 11027/14725 [5:36:48<1:52:13,  1.82s/it]
 75%|███████▍  | 11028/14725 [5:36:49<1:52:05,  1.82s/it]
 75
+0: {'loss': 0.7857, 'grad_norm': 0.6857631802501039, 'learning_rate': 2e-05, 'epoch': 0.75}
+0: %|███████▍  | 11029/14725 [5:36:51<1:50:40,  1.80s/it]
 75%|███████▍  | 11030/14725 [5:36:53<1:50:32,  1.79s/it]
                                                         

 75%|███████▍  | 11030/14725 [5:36:53<1:50:32,  1.79s/it]
 75%|███████▍  | 11031/14725 [5:36:55<1:50:41,  1.80s/it]
 75%|███████▍  | 11032/14725 [5:36:57<1:50:52,  1.80s/it]
 75%|███████▍  | 11033/14725 [5:36:58<1:50:11,  1.79s/it]
 75%|███████▍  | 11034/14725 [5:37:00<1:49:42,  1.78s/it]
 75%|███████▍  | 11035/14725 [5:37:02<1:49:36,  1.78s/it]
 75%|███████▍  | 11036/14725 [5:37:04<1:49:02,  1.77s/it]
 75%|███████▍  | 11037/14725 [5:37:06<1:50:10,  1.79s/it]
 75%|███████▍  | 11038/14725 [5:37:07<1:49:33,  1.78s/it]
 75%|███████▍  | 11039/14725 [5:37:09<1:50:00,  1.79s/it]
 75%|███████▍  | 11040/14725 [5:37:11<1:49:24,  1.78s/it]
      
+0: {'loss': 0.783, 'grad_norm': 0.7024259845676062, 'learning_rate': 2e-05, 'epoch': 0.75}
+0: {'loss': 0.8053, 'grad_norm': 0.6500272913099076, 'learning_rate': 2e-05, 'epoch': 0.75}
+0:                                                    

 75%|███████▍  | 11040/14725 [5:37:11<1:49:24,  1.78s/it]
 75%|███████▍  | 11041/14725 [5:37:13<1:49:25,  1.78s/it]
 75%|███████▍  | 11042/14725 [5:37:14<1:48:45,  1.77s/it]
 75%|███████▍  | 11043/14725 [5:37:16<1:50:13,  1.80s/it]
 75%|███████▌  | 11044/14725 [5:37:18<1:51:02,  1.81s/it]
 75%|███████▌  | 11045/14725 [5:37:20<1:51:23,  1.82s/it]
 75%|███████▌  | 11046/14725 [5:37:22<1:50:57,  1.81s/it]
 75%|███████▌  | 11047/14725 [5:37:24<1:50:57,  1.81s/it]
 75%|███████▌  | 11048/14725 [5:37:25<1:49:56,  1.79s/it]
 75%|███████▌  | 11049/14725 [5:37:27<1:51:18,  1.82s/it]
 75%|███████▌  | 11050/14725 [5:37:29<1:49:47,  1.79s/it]
                                                         

 75%|███████▌  | 11050/14725 [5:37:29<1:49:47,  1.79s/it]
 75%|██████�
+0: {'loss': 0.7962, 'grad_norm': 0.6415189110333996, 'learning_rate': 2e-05, 'epoch': 0.75}
+0: ��▌  | 11051/14725 [5:37:31<1:49:57,  1.80s/it]
 75%|███████▌  | 11052/14725 [5:37:33<1:51:46,  1.83s/it]
 75%|███████▌  | 11053/14725 [5:37:34<1:52:18,  1.83s/it]
 75%|███████▌  | 11054/14725 [5:37:36<1:52:21,  1.84s/it]
 75%|███████▌  | 11055/14725 [5:37:38<1:51:53,  1.83s/it]
 75%|███████▌  | 11056/14725 [5:37:40<1:54:58,  1.88s/it]
 75%|███████▌  | 11057/14725 [5:37:42<1:53:36,  1.86s/it]
 75%|███████▌  | 11058/14725 [5:37:44<1:53:49,  1.86s/it]
 75%|███████▌  | 11059/14725 [5:37:46<1:54:42,  1.88s/it]
 75%|███████▌  | 11060/14725 [5:37:48<1:53:46,  1.86s/it]
                                                         

 75%|███████▌  | 11060/14725 [5:37:48<1:53:46,  1.86s/it]
 75%|███████▌  | 11061/14725 [5:37:49<1:51:55,  1.83s/it]
 75%|███████▌  | 11062/14725 [5:37:51<1:50:35,  1.81s/it]
 75%|███████�
+0: {'loss': 0.7979, 'grad_norm': 0.6473372440895304, 'learning_rate': 2e-05, 'epoch': 0.75}
+0: ��  | 11063/14725 [5:37:53<1:51:00,  1.82s/it]
 75%|███████▌  | 11064/14725 [5:37:55<1:51:45,  1.83s/it]
 75%|███████▌  | 11065/14725 [5:37:56<1:50:30,  1.81s/it]
 75%|███████▌  | 11066/14725 [5:37:58<1:52:03,  1.84s/it]
 75%|███████▌  | 11067/14725 [5:38:00<1:50:36,  1.81s/it]
 75%|███████▌  | 11068/14725 [5:38:02<1:52:08,  1.84s/it]
 75%|███████▌  | 11069/14725 [5:38:04<1:51:46,  1.83s/it]
 75%|███████▌  | 11070/14725 [5:38:06<1:53:02,  1.86s/it]
                                                         

 75%|███████▌  | 11070/14725 [5:38:06<1:53:02,  1.86s/it]
 75%|███████▌  | 11071/14725 [5:38:08<1:52:10,  1.84s/it]
 75%|███████▌  | 11072/14725 [5:38:09<1:51:34,  1.83s/it]
 75%|███████▌  | 11073/14725 [5:38:11<1:49:54,  1.81s/it]
 75%|███████▌  | 11074/14725 [5:38:13<1:49:35,  1.80s/it]
 75%|███████▌ 
+0: {'loss': 0.7964, 'grad_norm': 0.6822704788449083, 'learning_rate': 2e-05, 'epoch': 0.75}
+0:  | 11075/14725 [5:38:15<1:48:53,  1.79s/it]
 75%|███████▌  | 11076/14725 [5:38:16<1:48:42,  1.79s/it]
 75%|███████▌  | 11077/14725 [5:38:18<1:48:04,  1.78s/it]
 75%|███████▌  | 11078/14725 [5:38:20<1:48:23,  1.78s/it]
 75%|███████▌  | 11079/14725 [5:38:22<1:48:38,  1.79s/it]
 75%|███████▌  | 11080/14725 [5:38:24<1:48:44,  1.79s/it]
                                                         

 75%|███████▌  | 11080/14725 [5:38:24<1:48:44,  1.79s/it]
 75%|███████▌  | 11081/14725 [5:38:25<1:50:00,  1.81s/it]
 75%|███████▌  | 11082/14725 [5:38:27<1:49:58,  1.81s/it]
 75%|███████▌  | 11083/14725 [5:38:29<1:49:21,  1.80s/it]
 75%|███████▌  | 11084/14725 [5:38:31<1:49:16,  1.80s/it]
 75%|███████▌  | 11085/14725 [5:38:33<1:49:02,  1.80s/it]
 75%|███████▌  | 11086/14725 [5:38:34<1:49:06,  1.80s/it]
 75%|███████▌  | 
+0: {'loss': 0.805, 'grad_norm': 0.6666371799022471, 'learning_rate': 2e-05, 'epoch': 0.75}
+0: 11087/14725 [5:38:36<1:49:38,  1.81s/it]
 75%|███████▌  | 11088/14725 [5:38:38<1:50:46,  1.83s/it]
 75%|███████▌  | 11089/14725 [5:38:40<1:51:00,  1.83s/it]
 75%|███████▌  | 11090/14725 [5:38:42<1:49:44,  1.81s/it]
                                                         

 75%|███████▌  | 11090/14725 [5:38:42<1:49:44,  1.81s/it]
 75%|███████▌  | 11091/14725 [5:38:44<1:49:10,  1.80s/it]
 75%|███████▌  | 11092/14725 [5:38:45<1:49:56,  1.82s/it]
 75%|███████▌  | 11093/14725 [5:38:47<1:49:35,  1.81s/it]
 75%|███████▌  | 11094/14725 [5:38:49<1:49:21,  1.81s/it]
 75%|███████▌  | 11095/14725 [5:38:51<1:49:42,  1.81s/it]
 75%|███████▌  | 11096/14725 [5:38:53<1:48:40,  1.80s/it]
 75%|███████▌  | 11097/14725 [5:38:54<1:47:40,  1.78s/it]
 75%|███████▌  | 11098/14725 [5:38:56<1:47:59,  1.79s/it]
 75%|███████▌  | 110
+0: {'loss': 0.7838, 'grad_norm': 0.7092954441681445, 'learning_rate': 2e-05, 'epoch': 0.75}
+0: 99/14725 [5:38:58<1:49:18,  1.81s/it]
 75%|███████▌  | 11100/14725 [5:39:00<1:50:10,  1.82s/it]
                                                         

 75%|███████▌  | 11100/14725 [5:39:00<1:50:10,  1.82s/it]
 75%|███████▌  | 11101/14725 [5:39:02<1:49:46,  1.82s/it]
 75%|███████▌  | 11102/14725 [5:39:03<1:48:57,  1.80s/it]
 75%|███████▌  | 11103/14725 [5:39:05<1:49:56,  1.82s/it]
 75%|███████▌  | 11104/14725 [5:39:07<1:49:04,  1.81s/it]
 75%|███████▌  | 11105/14725 [5:39:09<1:48:35,  1.80s/it]
 75%|███████▌  | 11106/14725 [5:39:11<1:47:57,  1.79s/it]
 75%|███████▌  | 11107/14725 [5:39:12<1:47:56,  1.79s/it]
 75%|███████▌  | 11108/14725 [5:39:14<1:48:00,  1.79s/it]
 75%|███████▌  | 11109/14725 [5:39:16<1:47:42,  1.79s/it]
 75%|███████▌  | 11110/14725 [5:39:18<1:49:48,  1.82s/it]
                                       
+0: {'loss': 0.8034, 'grad_norm': 0.6850022099739556, 'learning_rate': 2e-05, 'epoch': 0.75}
+0: {'loss': 0.8002, 'grad_norm': 0.6956193813981179, 'learning_rate': 2e-05, 'epoch': 0.76}
+0:                   

 75%|███████▌  | 11110/14725 [5:39:18<1:49:48,  1.82s/it]
 75%|███████▌  | 11111/14725 [5:39:20<1:50:29,  1.83s/it]
 75%|███████▌  | 11112/14725 [5:39:22<1:50:02,  1.83s/it]
 75%|███████▌  | 11113/14725 [5:39:23<1:48:40,  1.81s/it]
 75%|███████▌  | 11114/14725 [5:39:25<1:48:37,  1.80s/it]
 75%|███████▌  | 11115/14725 [5:39:27<1:48:23,  1.80s/it]
 75%|███████▌  | 11116/14725 [5:39:29<1:48:15,  1.80s/it]
 75%|███████▌  | 11117/14725 [5:39:31<1:50:52,  1.84s/it]
 76%|███████▌  | 11118/14725 [5:39:32<1:49:58,  1.83s/it]
 76%|███████▌  | 11119/14725 [5:39:34<1:48:43,  1.81s/it]
 76%|███████▌  | 11120/14725 [5:39:36<1:49:32,  1.82s/it]
                                                         

 76%|███████▌  | 11120/14725 [5:39:36<1:49:32,  1.82s/it]
 76%|███████▌  | 11121/14725 [5:39:38<1:4
+0: {'loss': 0.7836, 'grad_norm': 0.6595591887977232, 'learning_rate': 2e-05, 'epoch': 0.76}
+0: 9:15,  1.82s/it]
 76%|███████▌  | 11122/14725 [5:39:40<1:48:41,  1.81s/it]
 76%|███████▌  | 11123/14725 [5:39:42<1:49:35,  1.83s/it]
 76%|███████▌  | 11124/14725 [5:39:43<1:51:55,  1.87s/it]
 76%|███████▌  | 11125/14725 [5:39:45<1:51:55,  1.87s/it]
 76%|███████▌  | 11126/14725 [5:39:47<1:50:12,  1.84s/it]
 76%|███████▌  | 11127/14725 [5:39:49<1:49:53,  1.83s/it]
 76%|███████▌  | 11128/14725 [5:39:51<1:48:38,  1.81s/it]
 76%|███████▌  | 11129/14725 [5:39:53<1:48:43,  1.81s/it]
 76%|███████▌  | 11130/14725 [5:39:54<1:47:41,  1.80s/it]
                                                         

 76%|███████▌  | 11130/14725 [5:39:54<1:47:41,  1.80s/it]
 76%|███████▌  | 11131/14725 [5:39:56<1:47:55,  1.80s/it]
 76%|███████▌  | 11132/14725 [5:39:58<1:48:46,  1.82s/it]
 76%|███████▌  | 11133/14725 [5:40:00<1:48:4
+0: {'loss': 0.79, 'grad_norm': 0.6977527618301027, 'learning_rate': 2e-05, 'epoch': 0.76}
+0: 8,  1.82s/it]
 76%|███████▌  | 11134/14725 [5:40:02<1:47:42,  1.80s/it]
 76%|███████▌  | 11135/14725 [5:40:03<1:47:12,  1.79s/it]
 76%|███████▌  | 11136/14725 [5:40:05<1:46:49,  1.79s/it]
 76%|███████▌  | 11137/14725 [5:40:07<1:47:21,  1.80s/it]
 76%|███████▌  | 11138/14725 [5:40:09<1:46:50,  1.79s/it]
 76%|███████▌  | 11139/14725 [5:40:10<1:47:38,  1.80s/it]
 76%|███████▌  | 11140/14725 [5:40:12<1:47:39,  1.80s/it]
                                                         

 76%|███████▌  | 11140/14725 [5:40:12<1:47:39,  1.80s/it]
 76%|███████▌  | 11141/14725 [5:40:14<1:48:33,  1.82s/it]
 76%|███████▌  | 11142/14725 [5:40:16<1:47:49,  1.81s/it]
 76%|███████▌  | 11143/14725 [5:40:18<1:47:31,  1.80s/it]
 76%|███████▌  | 11144/14725 [5:40:20<1:49:15,  1.83s/it]
 76%|███████▌  | 11145/14725 [5:40:21<1:48:14, 
+0: {'loss': 0.7856, 'grad_norm': 0.6325516548631948, 'learning_rate': 2e-05, 'epoch': 0.76}
+0:  1.81s/it]
 76%|███████▌  | 11146/14725 [5:40:23<1:46:57,  1.79s/it]
 76%|███████▌  | 11147/14725 [5:40:25<1:47:14,  1.80s/it]
 76%|███████▌  | 11148/14725 [5:40:27<1:46:43,  1.79s/it]
 76%|███████▌  | 11149/14725 [5:40:28<1:45:52,  1.78s/it]
 76%|███████▌  | 11150/14725 [5:40:30<1:45:59,  1.78s/it]
                                                         

 76%|███████▌  | 11150/14725 [5:40:30<1:45:59,  1.78s/it]
 76%|███████▌  | 11151/14725 [5:40:32<1:46:15,  1.78s/it]
 76%|███████▌  | 11152/14725 [5:40:34<1:47:01,  1.80s/it]
 76%|███████▌  | 11153/14725 [5:40:36<1:47:47,  1.81s/it]
 76%|███████▌  | 11154/14725 [5:40:38<1:47:56,  1.81s/it]
 76%|███████▌  | 11155/14725 [5:40:39<1:48:32,  1.82s/it]
 76%|███████▌  | 11156/14725 [5:40:41<1:50:01,  1.85s/it]
 76%|███████▌  | 11157/14725 [5:40:43<1:48:06,  1.
+0: {'loss': 0.797, 'grad_norm': 0.7259832076163834, 'learning_rate': 2e-05, 'epoch': 0.76}
+0: 82s/it]
 76%|███████▌  | 11158/14725 [5:40:45<1:46:57,  1.80s/it]
 76%|███████▌  | 11159/14725 [5:40:47<1:46:22,  1.79s/it]
 76%|███████▌  | 11160/14725 [5:40:48<1:47:37,  1.81s/it]
                                                         

 76%|███████▌  | 11160/14725 [5:40:48<1:47:37,  1.81s/it]
 76%|███████▌  | 11161/14725 [5:40:50<1:47:14,  1.81s/it]
 76%|███████▌  | 11162/14725 [5:40:52<1:46:29,  1.79s/it]
 76%|███████▌  | 11163/14725 [5:40:54<1:46:43,  1.80s/it]
 76%|███████▌  | 11164/14725 [5:40:56<1:48:14,  1.82s/it]
 76%|███████▌  | 11165/14725 [5:40:58<1:49:43,  1.85s/it]
 76%|███████▌  | 11166/14725 [5:40:59<1:49:24,  1.84s/it]
 76%|███████▌  | 11167/14725 [5:41:01<1:48:17,  1.83s/it]
 76%|███████▌  | 11168/14725 [5:41:03<1:46:56,  1.80s/it]
 76%|███████▌  | 11169/14725 [5:41:05<1:47:19,  1.81s
+0: {'loss': 0.802, 'grad_norm': 0.7124455872054876, 'learning_rate': 2e-05, 'epoch': 0.76}
+0: {'loss': 0.7887, 'grad_norm': 0.6702939308470262, 'learning_rate': 2e-05, 'epoch': 0.76}
+0: /it]
 76%|███████▌  | 11170/14725 [5:41:07<1:46:57,  1.81s/it]
                                                         

 76%|███████▌  | 11170/14725 [5:41:07<1:46:57,  1.81s/it]
 76%|███████▌  | 11171/14725 [5:41:08<1:46:18,  1.79s/it]
 76%|███████▌  | 11172/14725 [5:41:10<1:45:45,  1.79s/it]
 76%|███████▌  | 11173/14725 [5:41:12<1:45:44,  1.79s/it]
 76%|███████▌  | 11174/14725 [5:41:14<1:45:11,  1.78s/it]
 76%|███████▌  | 11175/14725 [5:41:15<1:45:02,  1.78s/it]
 76%|███████▌  | 11176/14725 [5:41:17<1:45:08,  1.78s/it]
 76%|███████▌  | 11177/14725 [5:41:19<1:45:56,  1.79s/it]
 76%|███████▌  | 11178/14725 [5:41:21<1:46:52,  1.81s/it]
 76%|███████▌  | 11179/14725 [5:41:23<1:45:46,  1.79s/it]
 76%|███████▌  | 11180/14725 [5:41:24<1:45:00,  1.78s/it]
                                                         

 76%|██�
+0: {'loss': 0.7814, 'grad_norm': 0.6941997000713044, 'learning_rate': 2e-05, 'epoch': 0.76}
+0: �████▌  | 11180/14725 [5:41:24<1:45:00,  1.78s/it]
 76%|███████▌  | 11181/14725 [5:41:26<1:45:39,  1.79s/it]
 76%|███████▌  | 11182/14725 [5:41:28<1:45:33,  1.79s/it]
 76%|███████▌  | 11183/14725 [5:41:30<1:47:14,  1.82s/it]
 76%|███████▌  | 11184/14725 [5:41:32<1:46:19,  1.80s/it]
 76%|███████▌  | 11185/14725 [5:41:33<1:45:26,  1.79s/it]
 76%|███████▌  | 11186/14725 [5:41:35<1:45:36,  1.79s/it]
 76%|███████▌  | 11187/14725 [5:41:37<1:45:32,  1.79s/it]
 76%|███████▌  | 11188/14725 [5:41:39<1:45:06,  1.78s/it]
 76%|███████▌  | 11189/14725 [5:41:40<1:44:44,  1.78s/it]
 76%|███████▌  | 11190/14725 [5:41:42<1:45:10,  1.79s/it]
                                                         

 76%|███████▌  | 11190/14725 [5:41:42<1:45:10,  1.79s/it]
 76%|███████▌  | 11191/14725 [5:41:44<1:46:06,  1.80s/it]
 76%|███�
+0: {'loss': 0.7894, 'grad_norm': 0.673027156577391, 'learning_rate': 2e-05, 'epoch': 0.76}
+0: �███▌  | 11192/14725 [5:41:46<1:45:50,  1.80s/it]
 76%|███████▌  | 11193/14725 [5:41:48<1:45:23,  1.79s/it]
 76%|███████▌  | 11194/14725 [5:41:49<1:44:53,  1.78s/it]
 76%|███████▌  | 11195/14725 [5:41:51<1:44:54,  1.78s/it]
 76%|███████▌  | 11196/14725 [5:41:53<1:44:52,  1.78s/it]
 76%|███████▌  | 11197/14725 [5:41:55<1:44:54,  1.78s/it]
 76%|███████▌  | 11198/14725 [5:41:57<1:44:12,  1.77s/it]
 76%|███████▌  | 11199/14725 [5:41:58<1:44:36,  1.78s/it]
 76%|███████▌  | 11200/14725 [5:42:00<1:44:33,  1.78s/it]
                                                         

 76%|███████▌  | 11200/14725 [5:42:00<1:44:33,  1.78s/it]
 76%|███████▌  | 11201/14725 [5:42:02<1:45:32,  1.80s/it]
 76%|███████▌  | 11202/14725 [5:42:04<1:45:53,  1.80s/it]
 76%|███████▌  | 11203/14725 [5:42:06<1:45:54,  1.80s/it]
 76%|████�
+0: {'loss': 0.8036, 'grad_norm': 0.6830643174589338, 'learning_rate': 2e-05, 'epoch': 0.76}
+0: �██▌  | 11204/14725 [5:42:07<1:45:05,  1.79s/it]
 76%|███████▌  | 11205/14725 [5:42:09<1:46:10,  1.81s/it]
 76%|███████▌  | 11206/14725 [5:42:11<1:45:49,  1.80s/it]
 76%|███████▌  | 11207/14725 [5:42:13<1:44:34,  1.78s/it]
 76%|███████▌  | 11208/14725 [5:42:15<1:44:31,  1.78s/it]
 76%|███████▌  | 11209/14725 [5:42:16<1:45:48,  1.81s/it]
 76%|███████▌  | 11210/14725 [5:42:18<1:45:45,  1.81s/it]
                                                         

 76%|███████▌  | 11210/14725 [5:42:18<1:45:45,  1.81s/it]
 76%|███████▌  | 11211/14725 [5:42:20<1:48:29,  1.85s/it]
 76%|███████▌  | 11212/14725 [5:42:22<1:48:04,  1.85s/it]
 76%|███████▌  | 11213/14725 [5:42:24<1:46:13,  1.81s/it]
 76%|███████▌  | 11214/14725 [5:42:26<1:46:03,  1.81s/it]
 76%|███████▌  | 11215/14725 [5:42:27<1:45:14,  1.80s/it]
 76%|█████�
+0: {'loss': 0.7952, 'grad_norm': 0.692336518138521, 'learning_rate': 2e-05, 'epoch': 0.76}
+0: �█▌  | 11216/14725 [5:42:29<1:44:04,  1.78s/it]
 76%|███████▌  | 11217/14725 [5:42:31<1:44:04,  1.78s/it]
 76%|███████▌  | 11218/14725 [5:42:33<1:44:42,  1.79s/it]
 76%|███████▌  | 11219/14725 [5:42:34<1:45:57,  1.81s/it]
 76%|███████▌  | 11220/14725 [5:42:36<1:45:44,  1.81s/it]
                                                         

 76%|███████▌  | 11220/14725 [5:42:36<1:45:44,  1.81s/it]
 76%|█���█████▌  | 11221/14725 [5:42:38<1:46:06,  1.82s/it]
 76%|███████▌  | 11222/14725 [5:42:40<1:45:48,  1.81s/it]
 76%|███████▌  | 11223/14725 [5:42:42<1:45:09,  1.80s/it]
 76%|███████▌  | 11224/14725 [5:42:44<1:46:04,  1.82s/it]
 76%|███████▌  | 11225/14725 [5:42:45<1:46:03,  1.82s/it]
 76%|███████▌  | 11226/14725 [5:42:47<1:46:04,  1.82s/it]
 76%|███████▌  | 11227/14725 [5:42:49<1:46:11,  1.82s/it]
 76%|██████�
+0: {'loss': 0.8098, 'grad_norm': 0.663912252768429, 'learning_rate': 2e-05, 'epoch': 0.76}
+0: �▋  | 11228/14725 [5:42:51<1:47:29,  1.84s/it]
 76%|███████▋  | 11229/14725 [5:42:53<1:46:23,  1.83s/it]
 76%|███████▋  | 11230/14725 [5:42:55<1:46:47,  1.83s/it]
                                                         

 76%|███████▋  | 11230/14725 [5:42:55<1:46:47,  1.83s/it]
 76%|███████▋  | 11231/14725 [5:42:56<1:46:22,  1.83s/it]
 76%|███████▋  | 11232/14725 [5:42:58<1:46:43,  1.83s/it]
 76%|███████▋  | 11233/14725 [5:43:00<1:46:28,  1.83s/it]
 76%|███████▋  | 11234/14725 [5:43:02<1:44:58,  1.80s/it]
 76%|███████▋  | 11235/14725 [5:43:04<1:45:28,  1.81s/it]
 76%|███████▋  | 11236/14725 [5:43:05<1:44:27,  1.80s/it]
 76%|███████▋  | 11237/14725 [5:43:07<1:45:01,  1.81s/it]
 76%|███████▋  | 11238/14725 [5:43:09<1:44:33,  1.80s/it]
 76%|███████▋  | 11239/14725 [5:43:11<1:43:52,  1.79s/it]
 76%|███████�
+0: {'loss': 0.815, 'grad_norm': 0.6671594426548422, 'learning_rate': 2e-05, 'epoch': 0.76}
+0: {'loss': 0.8166, 'grad_norm': 0.700938913552597, 'learning_rate': 2e-05, 'epoch': 0.76}
+0: �  | 11240/14725 [5:43:13<1:44:35,  1.80s/it]
                                                         

 76%|███████▋  | 11240/14725 [5:43:13<1:44:35,  1.80s/it]
 76%|███████▋  | 11241/14725 [5:43:14<1:45:13,  1.81s/it]
 76%|███████▋  | 11242/14725 [5:43:16<1:46:10,  1.83s/it]
 76%|███████▋  | 11243/14725 [5:43:18<1:45:07,  1.81s/it]
 76%|███████▋  | 11244/14725 [5:43:20<1:46:18,  1.83s/it]
 76%|███████▋  | 11245/14725 [5:43:22<1:44:42,  1.81s/it]
 76%|███████▋  | 11246/14725 [5:43:23<1:43:52,  1.79s/it]
 76%|███████▋  | 11247/14725 [5:43:25<1:43:38,  1.79s/it]
 76%|███████▋  | 11248/14725 [5:43:27<1:44:05,  1.80s/it]
 76%|███████▋  | 11249/14725 [5:43:29<1:44:41,  1.81s/it]
 76%|███████▋  | 11250/14725 [5:43:31<1:44:40,  1.81s/it]
                                                         

 76%|███████▋  | 11250/14725 [
+0: {'loss': 0.8225, 'grad_norm': 0.7075999728450101, 'learning_rate': 2e-05, 'epoch': 0.76}
+0: 5:43:31<1:44:40,  1.81s/it]
 76%|███████▋  | 11251/14725 [5:43:32<1:44:34,  1.81s/it]
 76%|███████▋  | 11252/14725 [5:43:34<1:44:00,  1.80s/it]
 76%|███████▋  | 11253/14725 [5:43:36<1:44:53,  1.81s/it]
 76%|███████▋  | 11254/14725 [5:43:38<1:43:54,  1.80s/it]
 76%|███████▋  | 11255/14725 [5:43:40<1:43:04,  1.78s/it]
 76%|███████▋  | 11256/14725 [5:43:41<1:42:32,  1.77s/it]
 76%|███████▋  | 11257/14725 [5:43:43<1:42:29,  1.77s/it]
 76%|███████▋  | 11258/14725 [5:43:45<1:43:47,  1.80s/it]
 76%|███████▋  | 11259/14725 [5:43:47<1:43:11,  1.79s/it]
 76%|███████▋  | 11260/14725 [5:43:49<1:43:10,  1.79s/it]
                                                         

 76%|███████▋  | 11260/14725 [5:43:49<1:43:10,  1.79s/it]
 76%|███████▋  | 11261/14725 [5:43:50<1:43:10,  1.79s/it]
 76%|███████▋  | 11262/14725 [5:4
+0: {'loss': 0.7551, 'grad_norm': 0.6553029365881058, 'learning_rate': 2e-05, 'epoch': 0.77}
+0: 3:52<1:44:30,  1.81s/it]
 76%|███████▋  | 11263/14725 [5:43:54<1:43:53,  1.80s/it]
 76%|███████▋  | 11264/14725 [5:43:56<1:43:11,  1.79s/it]
 77%|███████▋  | 11265/14725 [5:43:58<1:43:40,  1.80s/it]
 77%|███████▋  | 11266/14725 [5:43:59<1:43:02,  1.79s/it]
 77%|███████▋  | 11267/14725 [5:44:01<1:42:23,  1.78s/it]
 77%|███████▋  | 11268/14725 [5:44:03<1:43:46,  1.80s/it]
 77%|███████▋  | 11269/14725 [5:44:05<1:44:42,  1.82s/it]
 77%|███████▋  | 11270/14725 [5:44:07<1:45:15,  1.83s/it]
                                                         

 77%|███████▋  | 11270/14725 [5:44:07<1:45:15,  1.83s/it]
 77%|███████▋  | 11271/14725 [5:44:08<1:45:21,  1.83s/it]
 77%|███████▋  | 11272/14725 [5:44:10<1:43:52,  1.80s/it]
 77%|███████▋  | 11273/14725 [5:44:12<1:43:57,  1.81s/it]
 77%|███████▋  | 11274/14725 [5:44:1
+0: {'loss': 0.7952, 'grad_norm': 0.7181820936467255, 'learning_rate': 2e-05, 'epoch': 0.77}
+0: 4<1:42:45,  1.79s/it]
 77%|███████▋  | 11275/14725 [5:44:16<1:42:21,  1.78s/it]
 77%|███████▋  | 11276/14725 [5:44:17<1:42:13,  1.78s/it]
 77%|███████▋  | 11277/14725 [5:44:19<1:43:00,  1.79s/it]
 77%|███████▋  | 11278/14725 [5:44:21<1:43:01,  1.79s/it]
 77%|███████▋  | 11279/14725 [5:44:23<1:44:02,  1.81s/it]
 77%|███████▋  | 11280/14725 [5:44:25<1:43:52,  1.81s/it]
                                                         

 77%|███████▋  | 11280/14725 [5:44:25<1:43:52,  1.81s/it]
 77%|███████▋  | 11281/14725 [5:44:26<1:43:20,  1.80s/it]
 77%|███████▋  | 11282/14725 [5:44:28<1:44:58,  1.83s/it]
 77%|███████▋  | 11283/14725 [5:44:30<1:43:59,  1.81s/it]
 77%|███████▋  | 11284/14725 [5:44:32<1:50:28,  1.93s/it]
 77%|███████▋  | 11285/14725 [5:44:34<1:48:25,  1.89s/it]
 77%|███████▋  | 11286/14725 [5:44:36<1
+0: {'loss': 0.7961, 'grad_norm': 0.675815317890735, 'learning_rate': 2e-05, 'epoch': 0.77}
+0: :47:09,  1.87s/it]
 77%|███████▋  | 11287/14725 [5:44:38<1:45:26,  1.84s/it]
 77%|███████▋  | 11288/14725 [5:44:39<1:44:27,  1.82s/it]
 77%|███████▋  | 11289/14725 [5:44:41<1:46:16,  1.86s/it]
 77%|███████▋  | 11290/14725 [5:44:43<1:44:44,  1.83s/it]
                                                         

 77%|███████▋  | 11290/14725 [5:44:43<1:44:44,  1.83s/it]
 77%|███████▋  | 11291/14725 [5:44:45<1:44:54,  1.83s/it]
 77%|███████▋  | 11292/14725 [5:44:47<1:44:35,  1.83s/it]
 77%|███████▋  | 11293/14725 [5:44:49<1:43:27,  1.81s/it]
 77%|███████▋  | 11294/14725 [5:44:50<1:42:36,  1.79s/it]
 77%|███████▋  | 11295/14725 [5:44:52<1:42:13,  1.79s/it]
 77%|███████▋  | 11296/14725 [5:44:54<1:42:56,  1.80s/it]
 77%|███████▋  | 11297/14725 [5:44:56<1:42:44,  1.80s/it]
 77%|███████▋  | 11298/14725 [5:44:57<1:43
+0: {'loss': 0.8001, 'grad_norm': 0.6401399938229586, 'learning_rate': 2e-05, 'epoch': 0.77}
+0: :15,  1.81s/it]
 77%|███████▋  | 11299/14725 [5:44:59<1:42:37,  1.80s/it]
 77%|███████▋  | 11300/14725 [5:45:01<1:47:39,  1.89s/it]
                                                         

 77%|███████▋  | 11300/14725 [5:45:01<1:47:39,  1.89s/it]
 77%|███████▋  | 11301/14725 [5:45:03<1:46:46,  1.87s/it]
 77%|███████▋  | 11302/14725 [5:45:05<1:45:04,  1.84s/it]
 77%|███████▋  | 11303/14725 [5:45:07<1:44:59,  1.84s/it]
 77%|███████▋  | 11304/14725 [5:45:09<1:43:33,  1.82s/it]
 77%|███████▋  | 11305/14725 [5:45:10<1:44:01,  1.82s/it]
 77%|███████▋  | 11306/14725 [5:45:12<1:42:40,  1.80s/it]
 77%|███████▋  | 11307/14725 [5:45:14<1:44:48,  1.84s/it]
 77%|███████▋  | 11308/14725 [5:45:16<1:44:17,  1.83s/it]
 77%|███████▋  | 11309/14725 [5:45:18<1:43:55,  1.83s/it]
 77%|███████▋  | 11310/14725 [5:45:20<1:44:34
+0: {'loss': 0.8172, 'grad_norm': 0.7005216568843766, 'learning_rate': 2e-05, 'epoch': 0.77}
+0: {'loss': 0.7962, 'grad_norm': 0.7142443629921365, 'learning_rate': 2e-05, 'epoch': 0.77}
+0: ,  1.84s/it]
                                                         

 77%|███████▋  | 11310/14725 [5:45:20<1:44:34,  1.84s/it]
 77%|███████▋  | 11311/14725 [5:45:21<1:43:41,  1.82s/it]
 77%|███████▋  | 11312/14725 [5:45:23<1:44:29,  1.84s/it]
 77%|███████▋  | 11313/14725 [5:45:25<1:44:38,  1.84s/it]
 77%|███████▋  | 11314/14725 [5:45:27<1:44:03,  1.83s/it]
 77%|███████▋  | 11315/14725 [5:45:29<1:45:43,  1.86s/it]
 77%|███████▋  | 11316/14725 [5:45:31<1:44:57,  1.85s/it]
 77%|███████▋  | 11317/14725 [5:45:32<1:44:34,  1.84s/it]
 77%|███████▋  | 11318/14725 [5:45:34<1:43:42,  1.83s/it]
 77%|███████▋  | 11319/14725 [5:45:36<1:43:36,  1.83s/it]
 77%|███████▋  | 11320/14725 [5:45:38<1:48:09,  1.91s/it]
                                                         

 77%|███████▋  | 11320/14725 [5:45:38<1:48:09,  1.91s/it]
 77%|
+0: {'loss': 0.799, 'grad_norm': 0.6402883702832274, 'learning_rate': 2e-05, 'epoch': 0.77}
+0: ███████▋  | 11321/14725 [5:45:40<1:45:51,  1.87s/it]
 77%|███████▋  | 11322/14725 [5:45:42<1:44:05,  1.84s/it]
 77%|███████▋  | 11323/14725 [5:45:44<1:43:37,  1.83s/it]
 77%|███████▋  | 11324/14725 [5:45:45<1:43:56,  1.83s/it]
 77%|███████▋  | 11325/14725 [5:45:47<1:43:01,  1.82s/it]
 77%|███████▋  | 11326/14725 [5:45:49<1:43:26,  1.83s/it]
 77%|██���████▋  | 11327/14725 [5:45:51<1:43:14,  1.82s/it]
 77%|███████▋  | 11328/14725 [5:45:53<1:42:17,  1.81s/it]
 77%|███████▋  | 11329/14725 [5:45:54<1:43:42,  1.83s/it]
 77%|███████▋  | 11330/14725 [5:45:56<1:42:39,  1.81s/it]
                                                         

 77%|███████▋  | 11330/14725 [5:45:56<1:42:39,  1.81s/it]
 77%|███████▋  | 11331/14725 [5:45:58<1:43:50,  1.84s/it]
 77%|███████▋  | 11332/14725 [5:46:00<1:45:37,  1.87s/it]
 77%|█
+0: {'loss': 0.7904, 'grad_norm': 0.6287262615353627, 'learning_rate': 2e-05, 'epoch': 0.77}
+0: ██████▋  | 11333/14725 [5:46:02<1:45:33,  1.87s/it]
 77%|███████▋  | 11334/14725 [5:46:04<1:43:55,  1.84s/it]
 77%|███████▋  | 11335/14725 [5:46:05<1:42:58,  1.82s/it]
 77%|███████▋  | 11336/14725 [5:46:07<1:42:28,  1.81s/it]
 77%|███████▋  | 11337/14725 [5:46:09<1:42:28,  1.81s/it]
 77%|███████▋  | 11338/14725 [5:46:11<1:48:02,  1.91s/it]
 77%|███████▋  | 11339/14725 [5:46:13<1:47:57,  1.91s/it]
 77%|███████▋  | 11340/14725 [5:46:15<1:47:01,  1.90s/it]
                                                         

 77%|███████▋  | 11340/14725 [5:46:15<1:47:01,  1.90s/it]
 77%|███████▋  | 11341/14725 [5:46:17<1:45:59,  1.88s/it]
 77%|███████▋  | 11342/14725 [5:46:19<1:46:44,  1.89s/it]
 77%|███████▋  | 11343/14725 [5:46:21<1:45:31,  1.87s/it]
 77%|███████▋  | 11344/14725 [5:46:22<1:44:52,  1.86s/it]
 77%|██
+0: {'loss': 0.7978, 'grad_norm': 0.6368161334758715, 'learning_rate': 2e-05, 'epoch': 0.77}
+0: █████▋  | 11345/14725 [5:46:24<1:43:51,  1.84s/it]
 77%|███████▋  | 11346/14725 [5:46:26<1:44:19,  1.85s/it]
 77%|███████▋  | 11347/14725 [5:46:28<1:42:52,  1.83s/it]
 77%|███████▋  | 11348/14725 [5:46:30<1:43:20,  1.84s/it]
 77%|███████▋  | 11349/14725 [5:46:32<1:43:16,  1.84s/it]
 77%|███████▋  | 11350/14725 [5:46:33<1:41:55,  1.81s/it]
                                                         

 77%|███████▋  | 11350/14725 [5:46:33<1:41:55,  1.81s/it]
 77%|███████▋  | 11351/14725 [5:46:35<1:42:36,  1.82s/it]
 77%|███████▋  | 11352/14725 [5:46:37<1:42:48,  1.83s/it]
 77%|███████▋  | 11353/14725 [5:46:39<1:42:30,  1.82s/it]
 77%|███████▋  | 11354/14725 [5:46:41<1:42:42,  1.83s/it]
 77%|███████▋  | 11355/14725 [5:46:42<1:42:18,  1.82s/it]
 77%|███████▋  | 11356/14725 [5:46:44<1:41:38,  1.81s/it]
 77%|███
+0: {'loss': 0.8113, 'grad_norm': 0.6849087895557937, 'learning_rate': 2e-05, 'epoch': 0.77}
+0: ████▋  | 11357/14725 [5:46:46<1:43:36,  1.85s/it]
 77%|███████▋  | 11358/14725 [5:46:48<1:42:39,  1.83s/it]
 77%|███████▋  | 11359/14725 [5:46:50<1:43:00,  1.84s/it]
 77%|███████▋  | 11360/14725 [5:46:52<1:42:35,  1.83s/it]
                                                         

 77%|███████▋  | 11360/14725 [5:46:52<1:42:35,  1.83s/it]
 77%|███████▋  | 11361/14725 [5:46:53<1:41:21,  1.81s/it]
 77%|███████▋  | 11362/14725 [5:46:55<1:42:00,  1.82s/it]
 77%|███████▋  | 11363/14725 [5:46:57<1:41:01,  1.80s/it]
 77%|███████▋  | 11364/14725 [5:46:59<1:40:28,  1.79s/it]
 77%|███████▋  | 11365/14725 [5:47:01<1:41:20,  1.81s/it]
 77%|███████▋  | 11366/14725 [5:47:02<1:41:31,  1.81s/it]
 77%|███████▋  | 11367/14725 [5:47:04<1:41:17,  1.81s/it]
 77%|███████▋  | 11368/14725 [5:47:06<1:41:13,  1.81s/it]
 77%|████
+0: {'loss': 0.8018, 'grad_norm': 0.6500022503437242, 'learning_rate': 2e-05, 'epoch': 0.77}
+0: ███▋  | 11369/14725 [5:47:08<1:41:02,  1.81s/it]
 77%|███████▋  | 11370/14725 [5:47:10<1:40:33,  1.80s/it]
                                                         

 77%|███████▋  | 11370/14725 [5:47:10<1:40:33,  1.80s/it]
 77%|███████▋  | 11371/14725 [5:47:11<1:40:01,  1.79s/it]
 77%|███████▋  | 11372/14725 [5:47:13<1:39:33,  1.78s/it]
 77%|███████▋  | 11373/14725 [5:47:15<1:43:31,  1.85s/it]
 77%|███████▋  | 11374/14725 [5:47:17<1:42:08,  1.83s/it]
 77%|███████▋  | 11375/14725 [5:47:19<1:42:29,  1.84s/it]
 77%|███████▋  | 11376/14725 [5:47:21<1:42:21,  1.83s/it]
 77%|███████▋  | 11377/14725 [5:47:22<1:40:57,  1.81s/it]
 77%|███████▋  | 11378/14725 [5:47:24<1:41:26,  1.82s/it]
 77%|███████▋  | 11379/14725 [5:47:26<1:41:36,  1.82s/it]
 77%|███████▋  | 11380/14725 [5:47:28<1:41:53,  1.83s/it]
                    
+0: {'loss': 0.8246, 'grad_norm': 0.6637427179876545, 'learning_rate': 2e-05, 'epoch': 0.77}
+0: {'loss': 0.7869, 'grad_norm': 0.7572360474318632, 'learning_rate': 2e-05, 'epoch': 0.77}
+0:                                      

 77%|███████▋  | 11380/14725 [5:47:28<1:41:53,  1.83s/it]
 77%|███████▋  | 11381/14725 [5:47:30<1:42:47,  1.84s/it]
 77%|███████▋  | 11382/14725 [5:47:32<1:41:48,  1.83s/it]
 77%|███████▋  | 11383/14725 [5:47:33<1:40:40,  1.81s/it]
 77%|███████▋  | 11384/14725 [5:47:35<1:40:09,  1.80s/it]
 77%|███████▋  | 11385/14725 [5:47:37<1:39:46,  1.79s/it]
 77%|███████▋  | 11386/14725 [5:47:39<1:39:23,  1.79s/it]
 77%|███████▋  | 11387/14725 [5:47:41<1:39:55,  1.80s/it]
 77%|███████▋  | 11388/14725 [5:47:42<1:39:24,  1.79s/it]
 77%|███████▋  | 11389/14725 [5:47:44<1:39:36,  1.79s/it]
 77%|███████▋  | 11390/14725 [5:47:46<1:38:45,  1.78s/it]
                                                         

 77%|███████▋  | 11390/14725 [5:47:46<1:38:45,  1.78s/it]
 77%|███████▋  | 11391
+0: {'loss': 0.7799, 'grad_norm': 0.7003337075363919, 'learning_rate': 2e-05, 'epoch': 0.77}
+0: /14725 [5:47:48<1:39:45,  1.80s/it]
 77%|███████▋  | 11392/14725 [5:47:49<1:39:40,  1.79s/it]
 77%|███████▋  | 11393/14725 [5:47:51<1:40:03,  1.80s/it]
 77%|███████▋  | 11394/14725 [5:47:53<1:39:55,  1.80s/it]
 77%|███████▋  | 11395/14725 [5:47:55<1:40:12,  1.81s/it]
 77%|███████▋  | 11396/14725 [5:47:57<1:41:27,  1.83s/it]
 77%|███████▋  | 11397/14725 [5:47:59<1:40:16,  1.81s/it]
 77%|███████▋  | 11398/14725 [5:48:00<1:40:06,  1.81s/it]
 77%|███████▋  | 11399/14725 [5:48:02<1:39:27,  1.79s/it]
 77%|███████▋  | 11400/14725 [5:48:04<1:38:57,  1.79s/it]
                                                         

 77%|███████▋  | 11400/14725 [5:48:04<1:38:57,  1.79s/it]
 77%|███████▋  | 11401/14725 [5:48:06<1:39:27,  1.80s/it]
 77%|███████▋  | 11402/14725 [5:48:07<1:39:31,  1.80s/it]
 77%|███████▋  | 11403/14
+0: {'loss': 0.7943, 'grad_norm': 0.6857389734523833, 'learning_rate': 2e-05, 'epoch': 0.77}
+0: 725 [5:48:09<1:39:08,  1.79s/it]
 77%|███████▋  | 11404/14725 [5:48:11<1:38:32,  1.78s/it]
 77%|███████▋  | 11405/14725 [5:48:13<1:37:58,  1.77s/it]
 77%|███████▋  | 11406/14725 [5:48:15<1:38:46,  1.79s/it]
 77%|███████▋  | 11407/14725 [5:48:16<1:40:46,  1.82s/it]
 77%|███████▋  | 11408/14725 [5:48:18<1:39:37,  1.80s/it]
 77%|███████▋  | 11409/14725 [5:48:20<1:38:27,  1.78s/it]
 77%|███████▋  | 11410/14725 [5:48:22<1:39:14,  1.80s/it]
                                                         

 77%|███████▋  | 11410/14725 [5:48:22<1:39:14,  1.80s/it]
 77%|███████▋  | 11411/14725 [5:48:24<1:39:19,  1.80s/it]
 78%|███████▊  | 11412/14725 [5:48:25<1:39:01,  1.79s/it]
 78%|███████▊  | 11413/14725 [5:48:27<1:38:30,  1.78s/it]
 78%|███████▊  | 11414/14725 [5:48:29<1:38:55,  1.79s/it]
 78%|███████▊  | 11415/14725
+0: {'loss': 0.7917, 'grad_norm': 0.6646207535434827, 'learning_rate': 2e-05, 'epoch': 0.78}
+0:  [5:48:31<1:40:27,  1.82s/it]
 78%|███████▊  | 11416/14725 [5:48:33<1:39:18,  1.80s/it]
 78%|███████▊  | 11417/14725 [5:48:34<1:39:36,  1.81s/it]
 78%|███████▊  | 11418/14725 [5:48:36<1:39:08,  1.80s/it]
 78%|███████▊  | 11419/14725 [5:48:38<1:38:30,  1.79s/it]
 78%|███████▊  | 11420/14725 [5:48:40<1:37:51,  1.78s/it]
                                                         

 78%|███████▊  | 11420/14725 [5:48:40<1:37:51,  1.78s/it]
 78%|███████▊  | 11421/14725 [5:48:42<1:37:57,  1.78s/it]
 78%|███████▊  | 11422/14725 [5:48:43<1:37:46,  1.78s/it]
 78%|███████▊  | 11423/14725 [5:48:45<1:38:40,  1.79s/it]
 78%|███████▊  | 11424/14725 [5:48:47<1:39:29,  1.81s/it]
 78%|███████▊  | 11425/14725 [5:48:49<1:39:45,  1.81s/it]
 78%|███████▊  | 11426/14725 [5:48:51<1:39:13,  1.80s/it]
 78%|███████▊  | 11427/14725 [5
+0: {'loss': 0.7956, 'grad_norm': 0.6701039494452564, 'learning_rate': 2e-05, 'epoch': 0.78}
+0: :48:52<1:39:11,  1.80s/it]
 78%|███████▊  | 11428/14725 [5:48:54<1:39:11,  1.81s/it]
 78%|███████▊  | 11429/14725 [5:48:56<1:38:47,  1.80s/it]
 78%|███████▊  | 11430/14725 [5:48:58<1:40:10,  1.82s/it]
                                                         

 78%|███████▊  | 11430/14725 [5:48:58<1:40:10,  1.82s/it]
 78%|███████▊  | 11431/14725 [5:49:00<1:40:19,  1.83s/it]
 78%|███████▊  | 11432/14725 [5:49:01<1:39:52,  1.82s/it]
 78%|███████▊  | 11433/14725 [5:49:03<1:39:48,  1.82s/it]
 78%|███████▊  | 11434/14725 [5:49:05<1:39:17,  1.81s/it]
 78%|███████▊  | 11435/14725 [5:49:07<1:39:09,  1.81s/it]
 78%|███████▊  | 11436/14725 [5:49:09<1:39:42,  1.82s/it]
 78%|███████▊  | 11437/14725 [5:49:11<1:39:25,  1.81s/it]
 78%|███████▊  | 11438/14725 [5:49:12<1:39:32,  1.82s/it]
 78%|███████▊  | 11439/14725 [5:49
+0: {'loss': 0.8018, 'grad_norm': 0.673974356694183, 'learning_rate': 2e-05, 'epoch': 0.78}
+0: :14<1:38:59,  1.81s/it]
 78%|███████▊  | 11440/14725 [5:49:16<1:38:50,  1.81s/it]
                                                         

 78%|███████▊  | 11440/14725 [5:49:16<1:38:50,  1.81s/it]
 78%|███████▊  | 11441/14725 [5:49:18<1:38:35,  1.80s/it]
 78%|███████▊  | 11442/14725 [5:49:20<1:38:29,  1.80s/it]
 78%|███████▊  | 11443/14725 [5:49:21<1:37:48,  1.79s/it]
 78%|███████▊  | 11444/14725 [5:49:23<1:39:00,  1.81s/it]
 78%|███████▊  | 11445/14725 [5:49:25<1:38:15,  1.80s/it]
 78%|███████▊  | 11446/14725 [5:49:27<1:37:32,  1.78s/it]
 78%|███████▊  | 11447/14725 [5:49:28<1:37:34,  1.79s/it]
 78%|███████▊  | 11448/14725 [5:49:30<1:37:08,  1.78s/it]
 78%|███████▊  | 11449/14725 [5:49:32<1:37:44,  1.79s/it]
 78%|███████▊  | 11450/14725 [5:49:34<1:37:07,  1.78s/it]
                                                     
+0: {'loss': 0.8053, 'grad_norm': 0.7369016943387332, 'learning_rate': 2e-05, 'epoch': 0.78}
+0: {'loss': 0.8171, 'grad_norm': 0.7113153320999256, 'learning_rate': 2e-05, 'epoch': 0.78}
+0:     

 78%|███████▊  | 11450/14725 [5:49:34<1:37:07,  1.78s/it]
 78%|███████▊  | 11451/14725 [5:49:36<1:36:51,  1.78s/it]
 78%|███████▊  | 11452/14725 [5:49:37<1:37:36,  1.79s/it]
 78%|███████▊  | 11453/14725 [5:49:39<1:38:17,  1.80s/it]
 78%|███████▊  | 11454/14725 [5:49:41<1:38:25,  1.81s/it]
 78%|███████▊  | 11455/14725 [5:49:43<1:37:48,  1.79s/it]
 78%|███████▊  | 11456/14725 [5:49:45<1:39:11,  1.82s/it]
 78%|███████▊  | 11457/14725 [5:49:47<1:40:08,  1.84s/it]
 78%|███████▊  | 11458/14725 [5:49:48<1:39:27,  1.83s/it]
 78%|███████▊  | 11459/14725 [5:49:50<1:38:37,  1.81s/it]
 78%|███████▊  | 11460/14725 [5:49:52<1:37:43,  1.80s/it]
                                                         

 78%|███████▊  | 11460/14725 [5:49:52<1:37:43,  1.80s/it]
 78%|███████▊  | 11461/14725 [5:49:54<1:37:22,  1.79s/i
+0: {'loss': 0.8087, 'grad_norm': 0.6891120439107081, 'learning_rate': 2e-05, 'epoch': 0.78}
+0: t]
 78%|███████▊  | 11462/14725 [5:49:56<1:38:51,  1.82s/it]
 78%|███████▊  | 11463/14725 [5:49:57<1:38:05,  1.80s/it]
 78%|███████▊  | 11464/14725 [5:49:59<1:38:00,  1.80s/it]
 78%|███████▊  | 11465/14725 [5:50:01<1:38:30,  1.81s/it]
 78%|███████▊  | 11466/14725 [5:50:03<1:38:23,  1.81s/it]
 78%|███████▊  | 11467/14725 [5:50:05<1:37:38,  1.80s/it]
 78%|███████▊  | 11468/14725 [5:50:06<1:37:02,  1.79s/it]
 78%|███████▊  | 11469/14725 [5:50:08<1:38:04,  1.81s/it]
 78%|███████▊  | 11470/14725 [5:50:10<1:37:45,  1.80s/it]
                                                         

 78%|███████▊  | 11470/14725 [5:50:10<1:37:45,  1.80s/it]
 78%|███████▊  | 11471/14725 [5:50:12<1:37:19,  1.79s/it]
 78%|███████▊  | 11472/14725 [5:50:14<1:37:47,  1.80s/it]
 78%|███████▊  | 11473/14725 [5:50:15<1:37:49,  1.80s/it]
+0: {'loss': 0.7817, 'grad_norm': 0.6466547520609348, 'learning_rate': 2e-05, 'epoch': 0.78}
+0:  78%|███████▊  | 11474/14725 [5:50:17<1:37:44,  1.80s/it]
 78%|███████▊  | 11475/14725 [5:50:19<1:37:15,  1.80s/it]
 78%|███████▊  | 11476/14725 [5:50:21<1:37:05,  1.79s/it]
 78%|███████▊  | 11477/14725 [5:50:22<1:36:30,  1.78s/it]
 78%|███████▊  | 11478/14725 [5:50:24<1:36:28,  1.78s/it]
 78%|███████▊  | 11479/14725 [5:50:26<1:36:37,  1.79s/it]
 78%|███████▊  | 11480/14725 [5:50:28<1:36:10,  1.78s/it]
                                                         

 78%|███████▊  | 11480/14725 [5:50:28<1:36:10,  1.78s/it]
 78%|███████▊  | 11481/14725 [5:50:30<1:35:51,  1.77s/it]
 78%|███████▊  | 11482/14725 [5:50:31<1:36:51,  1.79s/it]
 78%|███████▊  | 11483/14725 [5:50:33<1:37:41,  1.81s/it]
 78%|███████▊  | 11484/14725 [5:50:35<1:37:23,  1.80s/it]
 78%|███████▊  | 11485/14725 [5:50:37<1:37:42,  1.81s/it]
 78
+0: {'loss': 0.8012, 'grad_norm': 0.6511482933903986, 'learning_rate': 2e-05, 'epoch': 0.78}
+0: %|███████▊  | 11486/14725 [5:50:39<1:37:18,  1.80s/it]
 78%|███████▊  | 11487/14725 [5:50:41<1:38:18,  1.82s/it]
 78%|███████▊  | 11488/14725 [5:50:42<1:37:42,  1.81s/it]
 78%|███████▊  | 11489/14725 [5:50:44<1:36:52,  1.80s/it]
 78%|███████▊  | 11490/14725 [5:50:46<1:37:19,  1.81s/it]
                                                         

 78%|███████▊  | 11490/14725 [5:50:46<1:37:19,  1.81s/it]
 78%|███████▊  | 11491/14725 [5:50:48<1:37:24,  1.81s/it]
 78%|███████▊  | 11492/14725 [5:50:50<1:37:10,  1.80s/it]
 78%|███████▊  | 11493/14725 [5:50:51<1:36:23,  1.79s/it]
 78%|███████▊  | 11494/14725 [5:50:53<1:36:27,  1.79s/it]
 78%|███████▊  | 11495/14725 [5:50:55<1:36:17,  1.79s/it]
 78%|███████▊  | 11496/14725 [5:50:57<1:36:17,  1.79s/it]
 78%|███████▊  | 11497/14725 [5:50:58<1:36:50,  1.80s/it]
 78%|�
+0: {'loss': 0.7798, 'grad_norm': 0.6556976790647172, 'learning_rate': 2e-05, 'epoch': 0.78}
+0: ��██████▊  | 11498/14725 [5:51:00<1:36:51,  1.80s/it]
 78%|███████▊  | 11499/14725 [5:51:02<1:38:10,  1.83s/it]
 78%|███████▊  | 11500/14725 [5:51:04<1:37:45,  1.82s/it]
                                                         

 78%|███████▊  | 11500/14725 [5:51:04<1:37:45,  1.82s/it]
 78%|███████▊  | 11501/14725 [5:51:06<1:37:33,  1.82s/it]
 78%|███████▊  | 11502/14725 [5:51:08<1:37:02,  1.81s/it]
 78%|███████▊  | 11503/14725 [5:51:09<1:36:31,  1.80s/it]
 78%|███████▊  | 11504/14725 [5:51:11<1:36:25,  1.80s/it]
 78%|███████▊  | 11505/14725 [5:51:13<1:36:12,  1.79s/it]
 78%|███████▊  | 11506/14725 [5:51:15<1:36:36,  1.80s/it]
 78%|███████▊  | 11507/14725 [5:51:17<1:37:08,  1.81s/it]
 78%|███████▊  | 11508/14725 [5:51:18<1:36:50,  1.81s/it]
 78%|███████▊  | 11509/14725 [5:51:20<1:36:20,  1.80s/it]
 78%|█�
+0: {'loss': 0.7834, 'grad_norm': 0.671530997381256, 'learning_rate': 2e-05, 'epoch': 0.78}
+0: {'loss': 0.7875, 'grad_norm': 0.7023755845481544, 'learning_rate': 2e-05, 'epoch': 0.78}
+0: ��█████▊  | 11510/14725 [5:51:22<1:36:33,  1.80s/it]
                                                         

 78%|███████▊  | 11510/14725 [5:51:22<1:36:33,  1.80s/it]
 78%|███████▊  | 11511/14725 [5:51:24<1:36:01,  1.79s/it]
 78%|███████▊  | 11512/14725 [5:51:26<1:36:29,  1.80s/it]
 78%|███████▊  | 11513/14725 [5:51:27<1:37:27,  1.82s/it]
 78%|███████▊  | 11514/14725 [5:51:29<1:36:54,  1.81s/it]
 78%|███████▊  | 11515/14725 [5:51:31<1:36:13,  1.80s/it]
 78%|███████▊  | 11516/14725 [5:51:33<1:35:35,  1.79s/it]
 78%|███████▊  | 11517/14725 [5:51:35<1:36:21,  1.80s/it]
 78%|███████▊  | 11518/14725 [5:51:36<1:36:16,  1.80s/it]
 78%|███████▊  | 11519/14725 [5:51:38<1:36:57,  1.81s/it]
 78%|███████▊  | 11520/14725 [5:51:40<1:36:55,  1.81s/it]
                                                         

 78%|███████�
+0: {'loss': 0.8026, 'grad_norm': 0.6784930050484841, 'learning_rate': 2e-05, 'epoch': 0.78}
+0: ��  | 11520/14725 [5:51:40<1:36:55,  1.81s/it]
 78%|███████▊  | 11521/14725 [5:51:42<1:37:01,  1.82s/it]
 78%|███████▊  | 11522/14725 [5:51:44<1:36:41,  1.81s/it]
 78%|███████▊  | 11523/14725 [5:51:45<1:35:35,  1.79s/it]
 78%|███████▊  | 11524/14725 [5:51:47<1:34:57,  1.78s/it]
 78%|███████▊  | 11525/14725 [5:51:49<1:36:52,  1.82s/it]
 78%|███████▊  | 11526/14725 [5:51:51<1:37:21,  1.83s/it]
 78%|███████▊  | 11527/14725 [5:51:53<1:36:09,  1.80s/it]
 78%|███████▊  | 11528/14725 [5:51:54<1:35:37,  1.79s/it]
 78%|███████▊  | 11529/14725 [5:51:56<1:36:14,  1.81s/it]
 78%|███████▊  | 11530/14725 [5:51:58<1:35:34,  1.79s/it]
                                                         

 78%|███████▊  | 11530/14725 [5:51:58<1:35:34,  1.79s/it]
 78%|███████▊  | 11531/14725 [5:52:00<1:36:55,  1.82s/it]
 78%|███████▊ 
+0: {'loss': 0.7957, 'grad_norm': 0.6787869852525018, 'learning_rate': 2e-05, 'epoch': 0.78}
+0:  | 11532/14725 [5:52:02<1:36:46,  1.82s/it]
 78%|███████▊  | 11533/14725 [5:52:03<1:35:26,  1.79s/it]
 78%|███████▊  | 11534/14725 [5:52:05<1:36:15,  1.81s/it]
 78%|███████▊  | 11535/14725 [5:52:07<1:35:22,  1.79s/it]
 78%|███████▊  | 11536/14725 [5:52:09<1:34:54,  1.79s/it]
 78%|███████▊  | 11537/14725 [5:52:11<1:35:31,  1.80s/it]
 78%|███████▊  | 11538/14725 [5:52:12<1:35:03,  1.79s/it]
 78%|███████▊  | 11539/14725 [5:52:14<1:35:22,  1.80s/it]
 78%|███████▊  | 11540/14725 [5:52:16<1:36:13,  1.81s/it]
                                                         

 78%|███████▊  | 11540/14725 [5:52:16<1:36:13,  1.81s/it]
 78%|███████▊  | 11541/14725 [5:52:18<1:36:31,  1.82s/it]
 78%|███████▊  | 11542/14725 [5:52:20<1:35:56,  1.81s/it]
 78%|███████▊  | 11543/14725 [5:52:22<1:36:26,  1.82s/it]
 78%|███████▊  | 
+0: {'loss': 0.8026, 'grad_norm': 0.7224047929407929, 'learning_rate': 2e-05, 'epoch': 0.78}
+0: 11544/14725 [5:52:23<1:37:40,  1.84s/it]
 78%|███████▊  | 11545/14725 [5:52:25<1:38:14,  1.85s/it]
 78%|███████▊  | 11546/14725 [5:52:27<1:37:11,  1.83s/it]
 78%|███████▊  | 11547/14725 [5:52:29<1:36:00,  1.81s/it]
 78%|███████▊  | 11548/14725 [5:52:31<1:35:14,  1.80s/it]
 78%|███████▊  | 11549/14725 [5:52:32<1:35:13,  1.80s/it]
 78%|███████▊  | 11550/14725 [5:52:34<1:35:46,  1.81s/it]
                                                         

 78%|███████▊  | 11550/14725 [5:52:34<1:35:46,  1.81s/it]
 78%|███████▊  | 11551/14725 [5:52:36<1:36:15,  1.82s/it]
 78%|███████▊  | 11552/14725 [5:52:38<1:36:10,  1.82s/it]
 78%|███████▊  | 11553/14725 [5:52:40<1:36:59,  1.83s/it]
 78%|███████▊  | 11554/14725 [5:52:42<1:36:11,  1.82s/it]
 78%|███████▊  | 11555/14725 [5:52:43<1:35:22,  1.81s/it]
 78%|███████▊  | 115
+0: {'loss': 0.8163, 'grad_norm': 0.6938192296674623, 'learning_rate': 2e-05, 'epoch': 0.79}
+0: 56/14725 [5:52:45<1:34:22,  1.79s/it]
 78%|███████▊  | 11557/14725 [5:52:47<1:34:00,  1.78s/it]
 78%|███████▊  | 11558/14725 [5:52:49<1:33:31,  1.77s/it]
 78%|███████▊  | 11559/14725 [5:52:50<1:33:54,  1.78s/it]
 79%|███████▊  | 11560/14725 [5:52:52<1:34:40,  1.79s/it]
                                                         

 79%|███████▊  | 11560/14725 [5:52:52<1:34:40,  1.79s/it]
 79%|███████▊  | 11561/14725 [5:52:54<1:35:30,  1.81s/it]
 79%|███████▊  | 11562/14725 [5:52:56<1:35:45,  1.82s/it]
 79%|███████▊  | 11563/14725 [5:52:58<1:35:05,  1.80s/it]
 79%|███████▊  | 11564/14725 [5:53:00<1:35:24,  1.81s/it]
 79%|███████▊  | 11565/14725 [5:53:01<1:35:21,  1.81s/it]
 79%|███████▊  | 11566/14725 [5:53:03<1:34:42,  1.80s/it]
 79%|███████▊  | 11567/14725 [5:53:05<1:34:18,  1.79s/it]
 79%|███████▊  | 11568/
+0: {'loss': 0.7858, 'grad_norm': 0.736623756756867, 'learning_rate': 2e-05, 'epoch': 0.79}
+0: 14725 [5:53:07<1:35:57,  1.82s/it]
 79%|███████▊  | 11569/14725 [5:53:09<1:36:04,  1.83s/it]
 79%|███████▊  | 11570/14725 [5:53:10<1:35:13,  1.81s/it]
                                                         

 79%|███████▊  | 11570/14725 [5:53:10<1:35:13,  1.81s/it]
 79%|███████▊  | 11571/14725 [5:53:12<1:35:13,  1.81s/it]
 79%|███████▊  | 11572/14725 [5:53:14<1:35:53,  1.82s/it]
 79%|███████▊  | 11573/14725 [5:53:16<1:35:14,  1.81s/it]
 79%|███████▊  | 11574/14725 [5:53:18<1:34:42,  1.80s/it]
 79%|███████▊  | 11575/14725 [5:53:19<1:35:25,  1.82s/it]
 79%|███████▊  | 11576/14725 [5:53:21<1:35:57,  1.83s/it]
 79%|███████▊  | 11577/14725 [5:53:23<1:34:48,  1.81s/it]
 79%|███████▊  | 11578/14725 [5:53:25<1:35:22,  1.82s/it]
 79%|███████▊  | 11579/14725 [5:53:27<1:34:49,  1.81s/it]
 79%|███████▊  | 11580/147
+0: {'loss': 0.7923, 'grad_norm': 0.649995335437692, 'learning_rate': 2e-05, 'epoch': 0.79}
+0: {'loss': 0.7986, 'grad_norm': 0.7083480352652078, 'learning_rate': 2e-05, 'epoch': 0.79}
+0: 25 [5:53:29<1:35:34,  1.82s/it]
                                                         

 79%|███████▊  | 11580/14725 [5:53:29<1:35:34,  1.82s/it]
 79%|███████▊  | 11581/14725 [5:53:30<1:35:19,  1.82s/it]
 79%|███████▊  | 11582/14725 [5:53:32<1:34:27,  1.80s/it]
 79%|███████▊  | 11583/14725 [5:53:34<1:33:50,  1.79s/it]
 79%|███████▊  | 11584/14725 [5:53:36<1:33:29,  1.79s/it]
 79%|███████▊  | 11585/14725 [5:53:37<1:33:15,  1.78s/it]
 79%|███████▊  | 11586/14725 [5:53:39<1:35:25,  1.82s/it]
 79%|███████▊  | 11587/14725 [5:53:41<1:39:32,  1.90s/it]
 79%|███████▊  | 11588/14725 [5:53:43<1:37:36,  1.87s/it]
 79%|███████▊  | 11589/14725 [5:53:45<1:36:39,  1.85s/it]
 79%|███████▊  | 11590/14725 [5:53:47<1:35:18,  1.82s/it]
                                                         

 79%|███████▊  | 11590/14725 [5:53:47<1:35:1
+0: {'loss': 0.7927, 'grad_norm': 0.6802619866589864, 'learning_rate': 2e-05, 'epoch': 0.79}
+0: 8,  1.82s/it]
 79%|███████▊  | 11591/14725 [5:53:49<1:36:08,  1.84s/it]
 79%|███████▊  | 11592/14725 [5:53:50<1:35:29,  1.83s/it]
 79%|███████▊  | 11593/14725 [5:53:52<1:34:57,  1.82s/it]
 79%|███████▊  | 11594/14725 [5:53:54<1:34:41,  1.81s/it]
 79%|███████▊  | 11595/14725 [5:53:56<1:33:48,  1.80s/it]
 79%|███████▉  | 11596/14725 [5:53:58<1:34:13,  1.81s/it]
 79%|███████▉  | 11597/14725 [5:53:59<1:34:04,  1.80s/it]
 79%|███████▉  | 11598/14725 [5:54:01<1:34:04,  1.81s/it]
 79%|███████▉  | 11599/14725 [5:54:03<1:33:49,  1.80s/it]
 79%|███████▉  | 11600/14725 [5:54:05<1:34:08,  1.81s/it]
                                                         

 79%|███████▉  | 11600/14725 [5:54:05<1:34:08,  1.81s/it]
 79%|███████▉  | 11601/14725 [5:54:07<1:33:37,  1.80s/it]
 79%|███████▉  | 11602/14725 [5:54:08<1:33:04, 
+0: {'loss': 0.7982, 'grad_norm': 0.6897704564472935, 'learning_rate': 2e-05, 'epoch': 0.79}
+0:  1.79s/it]
 79%|███████▉  | 11603/14725 [5:54:10<1:32:15,  1.77s/it]
 79%|███████▉  | 11604/14725 [5:54:12<1:32:50,  1.78s/it]
 79%|███████▉  | 11605/14725 [5:54:14<1:33:08,  1.79s/it]
 79%|███████▉  | 11606/14725 [5:54:16<1:33:14,  1.79s/it]
 79%|███████▉  | 11607/14725 [5:54:17<1:33:18,  1.80s/it]
 79%|███████▉  | 11608/14725 [5:54:19<1:32:41,  1.78s/it]
 79%|███████▉  | 11609/14725 [5:54:21<1:33:14,  1.80s/it]
 79%|███████▉  | 11610/14725 [5:54:23<1:32:59,  1.79s/it]
                                                         

 79%|███████▉  | 11610/14725 [5:54:23<1:32:59,  1.79s/it]
 79%|███████▉  | 11611/14725 [5:54:25<1:32:36,  1.78s/it]
 79%|███████▉  | 11612/14725 [5:54:26<1:34:06,  1.81s/it]
 79%|███████▉  | 11613/14725 [5:54:28<1:35:03,  1.83s/it]
 79%|███████▉  | 11614/14725 [5:54:30<1:35:28,  1.
+0: {'loss': 0.8165, 'grad_norm': 0.675978678275415, 'learning_rate': 2e-05, 'epoch': 0.79}
+0: 84s/it]
 79%|███████▉  | 11615/14725 [5:54:32<1:34:03,  1.81s/it]
 79%|███████▉  | 11616/14725 [5:54:34<1:33:15,  1.80s/it]
 79%|███████▉  | 11617/14725 [5:54:36<1:33:52,  1.81s/it]
 79%|███████▉  | 11618/14725 [5:54:37<1:33:23,  1.80s/it]
 79%|███████▉  | 11619/14725 [5:54:39<1:34:16,  1.82s/it]
 79%|███████▉  | 11620/14725 [5:54:41<1:33:46,  1.81s/it]
                                                         

 79%|███████▉  | 11620/14725 [5:54:41<1:33:46,  1.81s/it]
 79%|███████▉  | 11621/14725 [5:54:43<1:34:14,  1.82s/it]
 79%|███████▉  | 11622/14725 [5:54:45<1:36:22,  1.86s/it]
 79%|███████▉  | 11623/14725 [5:54:47<1:35:21,  1.84s/it]
 79%|███████▉  | 11624/14725 [5:54:48<1:34:52,  1.84s/it]
 79%|███████▉  | 11625/14725 [5:54:50<1:34:32,  1.83s/it]
 79%|███████▉  | 11626/14725 [5:54:52<1:34:09,  1.82s
+0: {'loss': 0.8035, 'grad_norm': 0.6662275203869327, 'learning_rate': 2e-05, 'epoch': 0.79}
+0: /it]
 79%|███████▉  | 11627/14725 [5:54:54<1:35:12,  1.84s/it]
 79%|███████▉  | 11628/14725 [5:54:56<1:34:20,  1.83s/it]
 79%|███████▉  | 11629/14725 [5:54:57<1:34:10,  1.83s/it]
 79%|███████▉  | 11630/14725 [5:54:59<1:33:43,  1.82s/it]
                                                         

 79%|███████▉  | 11630/14725 [5:54:59<1:33:43,  1.82s/it]
 79%|███████▉  | 11631/14725 [5:55:01<1:32:43,  1.80s/it]
 79%|███████▉  | 11632/14725 [5:55:03<1:32:06,  1.79s/it]
 79%|███████▉  | 11633/14725 [5:55:05<1:31:30,  1.78s/it]
 79%|███████▉  | 11634/14725 [5:55:06<1:32:15,  1.79s/it]
 79%|███████▉  | 11635/14725 [5:55:08<1:31:45,  1.78s/it]
 79%|███████▉  | 11636/14725 [5:55:10<1:32:14,  1.79s/it]
 79%|███████▉  | 11637/14725 [5:55:12<1:32:06,  1.79s/it]
 79%|███████▉  | 11638/14725 [5:55:14<1:33:08,  1.81s/it
+0: {'loss': 0.7874, 'grad_norm': 0.6910203260400907, 'learning_rate': 2e-05, 'epoch': 0.79}
+0: ]
 79%|███████▉  | 11639/14725 [5:55:15<1:32:41,  1.80s/it]
 79%|███████▉  | 11640/14725 [5:55:17<1:31:57,  1.79s/it]
                                                         

 79%|███████▉  | 11640/14725 [5:55:17<1:31:57,  1.79s/it]
 79%|███████▉  | 11641/14725 [5:55:19<1:32:21,  1.80s/it]
 79%|███████▉  | 11642/14725 [5:55:21<1:31:42,  1.78s/it]
 79%|███████▉  | 11643/14725 [5:55:23<1:32:26,  1.80s/it]
 79%|███████▉  | 11644/14725 [5:55:24<1:32:19,  1.80s/it]
 79%|███████▉  | 11645/14725 [5:55:26<1:31:28,  1.78s/it]
 79%|███████▉  | 11646/14725 [5:55:28<1:31:17,  1.78s/it]
 79%|███████▉  | 11647/14725 [5:55:30<1:31:43,  1.79s/it]
 79%|███████▉  | 11648/14725 [5:55:31<1:31:19,  1.78s/it]
 79%|███████▉  | 11649/14725 [5:55:33<1:31:34,  1.79s/it]
 79%|███████▉  | 11650/14725 [5:55:35<1:31:45,  1.79s/it]
 
+0: {'loss': 0.7872, 'grad_norm': 0.6454126946908504, 'learning_rate': 2e-05, 'epoch': 0.79}
+0: {'loss': 0.8046, 'grad_norm': 0.6803976809072775, 'learning_rate': 2e-05, 'epoch': 0.79}
+0:                                                         

 79%|███████▉  | 11650/14725 [5:55:35<1:31:45,  1.79s/it]
 79%|███████▉  | 11651/14725 [5:55:37<1:31:59,  1.80s/it]
 79%|███████▉  | 11652/14725 [5:55:39<1:31:47,  1.79s/it]
 79%|███████▉  | 11653/14725 [5:55:40<1:31:58,  1.80s/it]
 79%|███████▉  | 11654/14725 [5:55:42<1:31:52,  1.79s/it]
 79%|███████▉  | 11655/14725 [5:55:44<1:32:29,  1.81s/it]
 79%|███████▉  | 11656/14725 [5:55:46<1:31:22,  1.79s/it]
 79%|███████▉  | 11657/14725 [5:55:48<1:32:34,  1.81s/it]
 79%|███████▉  | 11658/14725 [5:55:49<1:31:51,  1.80s/it]
 79%|███████▉  | 11659/14725 [5:55:51<1:33:16,  1.83s/it]
 79%|███████▉  | 11660/14725 [5:55:53<1:33:06,  1.82s/it]
                                                         

 79%|███████▉  | 11660/14725 [5:55:53<1:33:06,  1.82s/it]
 79%|████�
+0: {'loss': 0.8007, 'grad_norm': 0.7326880708655905, 'learning_rate': 2e-05, 'epoch': 0.79}
+0: �██▉  | 11661/14725 [5:55:55<1:33:14,  1.83s/it]
 79%|███████▉  | 11662/14725 [5:55:57<1:32:44,  1.82s/it]
 79%|███████▉  | 11663/14725 [5:55:59<1:32:43,  1.82s/it]
 79%|███████▉  | 11664/14725 [5:56:00<1:34:12,  1.85s/it]
 79%|███████▉  | 11665/14725 [5:56:02<1:33:29,  1.83s/it]
 79%|███████▉  | 11666/14725 [5:56:04<1:33:29,  1.83s/it]
 79%|███████▉  | 11667/14725 [5:56:06<1:33:10,  1.83s/it]
 79%|███████▉  | 11668/14725 [5:56:08<1:32:25,  1.81s/it]
 79%|███████▉  | 11669/14725 [5:56:09<1:31:51,  1.80s/it]
 79%|███████▉  | 11670/14725 [5:56:11<1:31:18,  1.79s/it]
                                                         

 79%|███████▉  | 11670/14725 [5:56:11<1:31:18,  1.79s/it]
 79%|███████▉  | 11671/14725 [5:56:13<1:32:54,  1.83s/it]
 79%|███████▉  | 11672/14725 [5:56:15<1:31:43,  1.80s/it]
 79%|█████�
+0: {'loss': 0.7918, 'grad_norm': 0.6820488476768334, 'learning_rate': 2e-05, 'epoch': 0.79}
+0: �█▉  | 11673/14725 [5:56:17<1:31:59,  1.81s/it]
 79%|███████▉  | 11674/14725 [5:56:19<1:31:56,  1.81s/it]
 79%|███████▉  | 11675/14725 [5:56:20<1:31:17,  1.80s/it]
 79%|███████▉  | 11676/14725 [5:56:22<1:31:20,  1.80s/it]
 79%|███████▉  | 11677/14725 [5:56:24<1:31:18,  1.80s/it]
 79%|███████▉  | 11678/14725 [5:56:26<1:30:59,  1.79s/it]
 79%|███████▉  | 11679/14725 [5:56:27<1:30:43,  1.79s/it]
 79%|███████▉  | 11680/14725 [5:56:29<1:31:20,  1.80s/it]
                                                         

 79%|███████▉  | 11680/14725 [5:56:29<1:31:20,  1.80s/it]
 79%|███████▉  | 11681/14725 [5:56:31<1:31:45,  1.81s/it]
 79%|███████▉  | 11682/14725 [5:56:33<1:31:53,  1.81s/it]
 79%|███████▉  | 11683/14725 [5:56:35<1:31:59,  1.81s/it]
 79%|███████▉  | 11684/14725 [5:56:37<1:33:01,  1.84s/it]
 79%|██████�
+0: {'loss': 0.7994, 'grad_norm': 0.6832578874174184, 'learning_rate': 2e-05, 'epoch': 0.79}
+0: �▉  | 11685/14725 [5:56:38<1:32:08,  1.82s/it]
 79%|███████▉  | 11686/14725 [5:56:40<1:32:16,  1.82s/it]
 79%|███████▉  | 11687/14725 [5:56:42<1:31:57,  1.82s/it]
 79%|███████▉  | 11688/14725 [5:56:44<1:31:59,  1.82s/it]
 79%|███████▉  | 11689/14725 [5:56:46<1:31:24,  1.81s/it]
 79%|███████▉  | 11690/14725 [5:56:47<1:31:00,  1.80s/it]
                                                         

 79%|███████▉  | 11690/14725 [5:56:47<1:31:00,  1.80s/it]
 79%|███████▉  | 11691/14725 [5:56:49<1:32:00,  1.82s/it]
 79%|███████▉  | 11692/14725 [5:56:51<1:32:16,  1.83s/it]
 79%|███████▉  | 11693/14725 [5:56:53<1:31:54,  1.82s/it]
 79%|███████▉  | 11694/14725 [5:56:55<1:32:16,  1.83s/it]
 79%|███████▉  | 11695/14725 [5:56:57<1:31:42,  1.82s/it]
 79%|███████▉  | 11696/14725 [5:56:58<1:30:38,  1.80s/it]
 79%|███████�
+0: {'loss': 0.7877, 'grad_norm': 0.7313917718697567, 'learning_rate': 2e-05, 'epoch': 0.79}
+0: �  | 11697/14725 [5:57:00<1:31:27,  1.81s/it]
 79%|███████▉  | 11698/14725 [5:57:02<1:31:59,  1.82s/it]
 79%|███████▉  | 11699/14725 [5:57:04<1:31:17,  1.81s/it]
 79%|███████▉  | 11700/14725 [5:57:06<1:30:21,  1.79s/it]
                                                         

 79%|███████▉  | 11700/14725 [5:57:06<1:30:21,  1.79s/it]
 79%|███████▉  | 11701/14725 [5:57:07<1:30:33,  1.80s/it]
 79%|███████▉  | 11702/14725 [5:57:09<1:31:24,  1.81s/it]
 79%|███████▉  | 11703/14725 [5:57:11<1:32:04,  1.83s/it]
 79%|███████▉  | 11704/14725 [5:57:13<1:31:43,  1.82s/it]
 79%|███████▉  | 11705/14725 [5:57:15<1:30:52,  1.81s/it]
 79%|███████▉  | 11706/14725 [5:57:17<1:31:26,  1.82s/it]
 80%|███████▉  | 11707/14725 [5:57:18<1:31:46,  1.82s/it]
 80%|███████▉  | 11708/14725 [5:57:20<1:32:31,  1.84s/it]
 80%|███████▉  
+0: {'loss': 0.7935, 'grad_norm': 0.669528234577385, 'learning_rate': 2e-05, 'epoch': 0.8}
+0: | 11709/14725 [5:57:22<1:31:05,  1.81s/it]
 80%|███████▉  | 11710/14725 [5:57:24<1:30:15,  1.80s/it]
                                                         

 80%|███████▉  | 11710/14725 [5:57:24<1:30:15,  1.80s/it]
 80%|███████▉  | 11711/14725 [5:57:26<1:30:53,  1.81s/it]
 80%|███████▉  | 11712/14725 [5:57:27<1:30:56,  1.81s/it]
 80%|███████▉  | 11713/14725 [5:57:29<1:31:14,  1.82s/it]
 80%|███████▉  | 11714/14725 [5:57:31<1:30:35,  1.81s/it]
 80%|███████▉  | 11715/14725 [5:57:33<1:29:37,  1.79s/it]
 80%|███████▉  | 11716/14725 [5:57:35<1:29:30,  1.78s/it]
 80%|███████▉  | 11717/14725 [5:57:36<1:30:39,  1.81s/it]
 80%|███████▉  | 11718/14725 [5:57:38<1:31:11,  1.82s/it]
 80%|███████▉  | 11719/14725 [5:57:40<1:31:41,  1.83s/it]
 80%|███████▉  | 11720/14725 [5:57:42<1:31:19,  1.82s/it]
                                  
+0: {'loss': 0.7952, 'grad_norm': 0.695154373785147, 'learning_rate': 2e-05, 'epoch': 0.8}
+0: {'loss': 0.7994, 'grad_norm': 0.7191677252507056, 'learning_rate': 2e-05, 'epoch': 0.8}
+0:                        

 80%|███████▉  | 11720/14725 [5:57:42<1:31:19,  1.82s/it]
 80%|███████▉  | 11721/14725 [5:57:44<1:31:21,  1.82s/it]
 80%|███████▉  | 11722/14725 [5:57:46<1:31:07,  1.82s/it]
 80%|███████▉  | 11723/14725 [5:57:47<1:30:00,  1.80s/it]
 80%|███████▉  | 11724/14725 [5:57:49<1:29:38,  1.79s/it]
 80%|███████▉  | 11725/14725 [5:57:51<1:31:01,  1.82s/it]
 80%|███████▉  | 11726/14725 [5:57:53<1:30:55,  1.82s/it]
 80%|███████▉  | 11727/14725 [5:57:55<1:31:09,  1.82s/it]
 80%|███████▉  | 11728/14725 [5:57:56<1:31:04,  1.82s/it]
 80%|███████▉  | 11729/14725 [5:57:58<1:30:35,  1.81s/it]
 80%|███████▉  | 11730/14725 [5:58:00<1:29:36,  1.80s/it]
                                                         

 80%|███████▉  | 11730/14725 [5:58:00<1:29:36,  1.80s/it]
 80%|███████▉  | 11731/14725 [5:58:0
+0: {'loss': 0.7944, 'grad_norm': 0.7502196641015251, 'learning_rate': 2e-05, 'epoch': 0.8}
+0: 2<1:30:29,  1.81s/it]
 80%|███████▉  | 11732/14725 [5:58:04<1:29:52,  1.80s/it]
 80%|███████▉  | 11733/14725 [5:58:05<1:29:27,  1.79s/it]
 80%|███████▉  | 11734/14725 [5:58:07<1:31:11,  1.83s/it]
 80%|███████▉  | 11735/14725 [5:58:09<1:31:02,  1.83s/it]
 80%|███████▉  | 11736/14725 [5:58:11<1:30:21,  1.81s/it]
 80%|███████▉  | 11737/14725 [5:58:13<1:29:56,  1.81s/it]
 80%|███████▉  | 11738/14725 [5:58:15<1:30:54,  1.83s/it]
 80%|███████▉  | 11739/14725 [5:58:16<1:30:04,  1.81s/it]
 80%|███████▉  | 11740/14725 [5:58:18<1:30:35,  1.82s/it]
                                                         

 80%|███████▉  | 11740/14725 [5:58:18<1:30:35,  1.82s/it]
 80%|███████▉  | 11741/14725 [5:58:20<1:30:35,  1.82s/it]
 80%|███████▉  | 11742/14725 [5:58:22<1:29:30,  1.80s/it]
 80%|███████▉  | 11743/14725 [5:58:24<1
+0: {'loss': 0.7823, 'grad_norm': 0.7111736085769924, 'learning_rate': 2e-05, 'epoch': 0.8}
+0: :29:47,  1.81s/it]
 80%|███████▉  | 11744/14725 [5:58:25<1:31:42,  1.85s/it]
 80%|███████▉  | 11745/14725 [5:58:27<1:30:57,  1.83s/it]
 80%|███████▉  | 11746/14725 [5:58:29<1:31:37,  1.85s/it]
 80%|███████▉  | 11747/14725 [5:58:31<1:30:52,  1.83s/it]
 80%|███████▉  | 11748/14725 [5:58:33<1:30:40,  1.83s/it]
 80%|███████▉  | 11749/14725 [5:58:35<1:32:42,  1.87s/it]
 80%|█��█████▉  | 11750/14725 [5:58:37<1:32:15,  1.86s/it]
                                                         

 80%|███████▉  | 11750/14725 [5:58:37<1:32:15,  1.86s/it]
 80%|███████▉  | 11751/14725 [5:58:38<1:30:56,  1.83s/it]
 80%|███████▉  | 11752/14725 [5:58:40<1:30:29,  1.83s/it]
 80%|███████▉  | 11753/14725 [5:58:42<1:29:16,  1.80s/it]
 80%|███████▉  | 11754/14725 [5:58:44<1:30:43,  1.83s/it]
 80%|███████▉  | 11755/14725 [5:58:46<1:29
+0: {'loss': 0.796, 'grad_norm': 0.716028983476549, 'learning_rate': 2e-05, 'epoch': 0.8}
+0: :59,  1.82s/it]
 80%|███████▉  | 11756/14725 [5:58:47<1:29:39,  1.81s/it]
 80%|███████▉  | 11757/14725 [5:58:49<1:29:17,  1.81s/it]
 80%|███████▉  | 11758/14725 [5:58:51<1:29:25,  1.81s/it]
 80%|███████▉  | 11759/14725 [5:58:53<1:30:02,  1.82s/it]
 80%|███████▉  | 11760/14725 [5:58:55<1:29:54,  1.82s/it]
                                                         

 80%|███████▉  | 11760/14725 [5:58:55<1:29:54,  1.82s/it]
 80%|███████▉  | 11761/14725 [5:58:57<1:30:45,  1.84s/it]
 80%|███████▉  | 11762/14725 [5:58:58<1:30:05,  1.82s/it]
 80%|███████▉  | 11763/14725 [5:59:00<1:29:00,  1.80s/it]
 80%|███████▉  | 11764/14725 [5:59:02<1:30:09,  1.83s/it]
 80%|███████▉  | 11765/14725 [5:59:04<1:29:19,  1.81s/it]
 80%|███████▉  | 11766/14725 [5:59:06<1:29:04,  1.81s/it]
 80%|███████▉  | 11767/14725 [5:59:07<1:28:32
+0: {'loss': 0.7852, 'grad_norm': 0.6560478212730976, 'learning_rate': 2e-05, 'epoch': 0.8}
+0: ,  1.80s/it]
 80%|███████▉  | 11768/14725 [5:59:09<1:27:41,  1.78s/it]
 80%|███████▉  | 11769/14725 [5:59:11<1:28:30,  1.80s/it]
 80%|███████▉  | 11770/14725 [5:59:13<1:28:59,  1.81s/it]
                                                         

 80%|███████▉  | 11770/14725 [5:59:13<1:28:59,  1.81s/it]
 80%|███████▉  | 11771/14725 [5:59:15<1:29:00,  1.81s/it]
 80%|███████▉  | 11772/14725 [5:59:16<1:28:01,  1.79s/it]
 80%|███████▉  | 11773/14725 [5:59:18<1:28:23,  1.80s/it]
 80%|███████▉  | 11774/14725 [5:59:20<1:28:14,  1.79s/it]
 80%|███████▉  | 11775/14725 [5:59:22<1:28:12,  1.79s/it]
 80%|███████▉  | 11776/14725 [5:59:24<1:28:41,  1.80s/it]
 80%|███████▉  | 11777/14725 [5:59:25<1:29:41,  1.83s/it]
 80%|███████▉  | 11778/14725 [5:59:27<1:28:32,  1.80s/it]
 80%|███████▉  | 11779/14725 [5:59:29<1:29:08,  
+0: {'loss': 0.788, 'grad_norm': 0.6638970726006371, 'learning_rate': 2e-05, 'epoch': 0.8}
+0: {'loss': 0.7885, 'grad_norm': 0.7166947367422678, 'learning_rate': 2e-05, 'epoch': 0.8}
+0: 1.82s/it]
 80%|████████  | 11780/14725 [5:59:31<1:28:36,  1.81s/it]
                                                         

 80%|████████  | 11780/14725 [5:59:31<1:28:36,  1.81s/it]
 80%|████████  | 11781/14725 [5:59:36<2:12:46,  2.71s/it]
 80%|████████  | 11782/14725 [5:59:37<2:00:13,  2.45s/it]
 80%|████████  | 11783/14725 [5:59:39<1:50:49,  2.26s/it]
 80%|████████  | 11784/14725 [5:59:41<1:44:31,  2.13s/it]
 80%|████████  | 11785/14725 [5:59:43<1:39:06,  2.02s/it]
 80%|████████  | 11786/14725 [5:59:45<1:36:30,  1.97s/it]
 80%|████████  | 11787/14725 [5:59:46<1:33:46,  1.92s/it]
 80%|████████  | 11788/14725 [5:59:48<1:31:57,  1.88s/it]
 80%|████████  | 11789/14725 [5:59:50<1:31:21,  1.87s/it]
 80%|████████  | 11790/14725 [5:59:52<1:30:12,  1.84s/it]
                                                         

 80%|█
+0: {'loss': 0.8, 'grad_norm': 0.6430820845374252, 'learning_rate': 2e-05, 'epoch': 0.8}
+0: ███████  | 11790/14725 [5:59:52<1:30:12,  1.84s/it]
 80%|████████  | 11791/14725 [5:59:54<1:30:18,  1.85s/it]
 80%|████████  | 11792/14725 [5:59:56<1:29:24,  1.83s/it]
 80%|████████  | 11793/14725 [5:59:57<1:28:44,  1.82s/it]
 80%|████████  | 11794/14725 [5:59:59<1:27:53,  1.80s/it]
 80%|████████  | 11795/14725 [6:00:01<1:28:18,  1.81s/it]
 80%|████████  | 11796/14725 [6:00:03<1:29:02,  1.82s/it]
 80%|████████  | 11797/14725 [6:00:05<1:28:20,  1.81s/it]
 80%|████████  | 11798/14725 [6:00:06<1:27:58,  1.80s/it]
 80%|████████  | 11799/14725 [6:00:08<1:28:36,  1.82s/it]
 80%|████████  | 11800/14725 [6:00:10<1:29:43,  1.84s/it]
                                                         

 80%|████████  | 11800/14725 [6:00:10<1:29:43,  1.84s/it]
 80%|████████  | 11801/14725 [6:00:12<1:30:39,  1.86s/it]
 80%|█��
+0: {'loss': 0.8006, 'grad_norm': 0.6508950560159714, 'learning_rate': 2e-05, 'epoch': 0.8}
+0: ██████  | 11802/14725 [6:00:14<1:29:46,  1.84s/it]
 80%|████████  | 11803/14725 [6:00:16<1:28:34,  1.82s/it]
 80%|████████  | 11804/14725 [6:00:17<1:27:44,  1.80s/it]
 80%|████████  | 11805/14725 [6:00:19<1:27:29,  1.80s/it]
 80%|████████  | 11806/14725 [6:00:21<1:26:52,  1.79s/it]
 80%|████████  | 11807/14725 [6:00:23<1:27:02,  1.79s/it]
 80%|████████  | 11808/14725 [6:00:25<1:28:22,  1.82s/it]
 80%|████████  | 11809/14725 [6:00:26<1:28:17,  1.82s/it]
 80%|████████  | 11810/14725 [6:00:28<1:27:16,  1.80s/it]
                                                         

 80%|████████  | 11810/14725 [6:00:28<1:27:16,  1.80s/it]
 80%|████████  | 11811/14725 [6:00:30<1:27:51,  1.81s/it]
 80%|████████  | 11812/14725 [6:00:32<1:27:46,  1.81s/it]
 80%|████████  | 11813/14725 [6:00:34<1:28:49,  1.83s/it]
 80%|███
+0: {'loss': 0.8071, 'grad_norm': 0.6224806904230642, 'learning_rate': 2e-05, 'epoch': 0.8}
+0: █████  | 11814/14725 [6:00:35<1:28:09,  1.82s/it]
 80%|████████  | 11815/14725 [6:00:37<1:27:34,  1.81s/it]
 80%|████████  | 11816/14725 [6:00:39<1:27:32,  1.81s/it]
 80%|████████  | 11817/14725 [6:00:41<1:26:37,  1.79s/it]
 80%|████████  | 11818/14725 [6:00:43<1:26:29,  1.79s/it]
 80%|████████  | 11819/14725 [6:00:44<1:26:49,  1.79s/it]
 80%|████████  | 11820/14725 [6:00:46<1:26:02,  1.78s/it]
                                                         

 80%|████████  | 11820/14725 [6:00:46<1:26:02,  1.78s/it]
 80%|████████  | 11821/14725 [6:00:48<1:26:23,  1.79s/it]
 80%|████████  | 11822/14725 [6:00:50<1:26:59,  1.80s/it]
 80%|████████  | 11823/14725 [6:00:51<1:26:29,  1.79s/it]
 80%|████████  | 11824/14725 [6:00:53<1:27:28,  1.81s/it]
 80%|████████  | 11825/14725 [6:00:55<1:26:32,  1.79s/it]
 80%|████
+0: {'loss': 0.8009, 'grad_norm': 0.6953662314071735, 'learning_rate': 2e-05, 'epoch': 0.8}
+0: ████  | 11826/14725 [6:00:57<1:26:56,  1.80s/it]
 80%|████████  | 11827/14725 [6:00:59<1:27:22,  1.81s/it]
 80%|████████  | 11828/14725 [6:01:01<1:26:57,  1.80s/it]
 80%|████████  | 11829/14725 [6:01:02<1:27:13,  1.81s/it]
 80%|████████  | 11830/14725 [6:01:04<1:27:40,  1.82s/it]
                                                         

 80%|████████  | 11830/14725 [6:01:04<1:27:40,  1.82s/it]
 80%|████████  | 11831/14725 [6:01:06<1:27:06,  1.81s/it]
 80%|████████  | 11832/14725 [6:01:08<1:28:20,  1.83s/it]
 80%|████████  | 11833/14725 [6:01:10<1:27:44,  1.82s/it]
 80%|████████  | 11834/14725 [6:01:11<1:27:19,  1.81s/it]
 80%|████████  | 11835/14725 [6:01:13<1:26:25,  1.79s/it]
 80%|████████  | 11836/14725 [6:01:15<1:27:52,  1.83s/it]
 80%|████████  | 11837/14725 [6:01:17<1:27:29,  1.82s/it]
 80%|█████
+0: {'loss': 0.7958, 'grad_norm': 0.6650034087377957, 'learning_rate': 2e-05, 'epoch': 0.8}
+0: ███  | 11838/14725 [6:01:19<1:27:08,  1.81s/it]
 80%|████████  | 11839/14725 [6:01:20<1:26:59,  1.81s/it]
 80%|████████  | 11840/14725 [6:01:22<1:26:01,  1.79s/it]
                                                         

 80%|████████  | 11840/14725 [6:01:22<1:26:01,  1.79s/it]
 80%|████████  | 11841/14725 [6:01:24<1:25:51,  1.79s/it]
 80%|████████  | 11842/14725 [6:01:26<1:26:02,  1.79s/it]
 80%|████████  | 11843/14725 [6:01:28<1:26:00,  1.79s/it]
 80%|████████  | 11844/14725 [6:01:29<1:26:51,  1.81s/it]
 80%|████████  | 11845/14725 [6:01:31<1:26:07,  1.79s/it]
 80%|████████  | 11846/14725 [6:01:33<1:26:53,  1.81s/it]
 80%|████████  | 11847/14725 [6:01:35<1:26:16,  1.80s/it]
 80%|████████  | 11848/14725 [6:01:37<1:26:54,  1.81s/it]
 80%|████████  | 11849/14725 [6:01:39<1:28:04,  1.84s/it]
 80%|██████
+0: {'loss': 0.7866, 'grad_norm': 0.6625083558492727, 'learning_rate': 2e-05, 'epoch': 0.8}
+0: {'loss': 0.7915, 'grad_norm': 0.6775024756889965, 'learning_rate': 2e-05, 'epoch': 0.81}
+0: ██  | 11850/14725 [6:01:40<1:27:18,  1.82s/it]
                                                         

 80%|████████  | 11850/14725 [6:01:40<1:27:18,  1.82s/it]
 80%|████████  | 11851/14725 [6:01:42<1:26:58,  1.82s/it]
 80%|████████  | 11852/14725 [6:01:44<1:26:33,  1.81s/it]
 80%|████████  | 11853/14725 [6:01:46<1:25:44,  1.79s/it]
 81%|████████  | 11854/14725 [6:01:48<1:26:06,  1.80s/it]
 81%|████████  | 11855/14725 [6:01:49<1:26:21,  1.81s/it]
 81%|████████  | 11856/14725 [6:01:51<1:26:41,  1.81s/it]
 81%|████████  | 11857/14725 [6:01:53<1:26:04,  1.80s/it]
 81%|████████  | 11858/14725 [6:01:55<1:25:24,  1.79s/it]
 81%|████████  | 11859/14725 [6:01:56<1:25:08,  1.78s/it]
 81%|████████  | 11860/14725 [6:01:58<1:24:52,  1.78s/it]
                                                         

 81%|████████  | 11860/14
+0: {'loss': 0.7941, 'grad_norm': 0.7027307714586809, 'learning_rate': 2e-05, 'epoch': 0.81}
+0: 725 [6:01:58<1:24:52,  1.78s/it]
 81%|████████  | 11861/14725 [6:02:00<1:24:51,  1.78s/it]
 81%|████████  | 11862/14725 [6:02:02<1:24:42,  1.78s/it]
 81%|████████  | 11863/14725 [6:02:04<1:24:28,  1.77s/it]
 81%|████████  | 11864/14725 [6:02:05<1:24:04,  1.76s/it]
 81%|████████  | 11865/14725 [6:02:07<1:24:18,  1.77s/it]
 81%|████████  | 11866/14725 [6:02:09<1:24:24,  1.77s/it]
 81%|████████  | 11867/14725 [6:02:11<1:24:34,  1.78s/it]
 81%|████████  | 11868/14725 [6:02:12<1:24:24,  1.77s/it]
 81%|████████  | 11869/14725 [6:02:14<1:25:24,  1.79s/it]
 81%|████████  | 11870/14725 [6:02:16<1:25:37,  1.80s/it]
                                                         

 81%|████████  | 11870/14725 [6:02:16<1:25:37,  1.80s/it]
 81%|████████  | 11871/14725 [6:02:18<1:26:20,  1.82s/it]
 81%|████████  | 11872/14725
+0: {'loss': 0.7848, 'grad_norm': 0.6641691237848678, 'learning_rate': 2e-05, 'epoch': 0.81}
+0:  [6:02:20<1:26:02,  1.81s/it]
 81%|████████  | 11873/14725 [6:02:22<1:26:55,  1.83s/it]
 81%|████████  | 11874/14725 [6:02:23<1:26:43,  1.83s/it]
 81%|████████  | 11875/14725 [6:02:25<1:25:26,  1.80s/it]
 81%|████████  | 11876/14725 [6:02:27<1:26:33,  1.82s/it]
 81%|████████  | 11877/14725 [6:02:29<1:26:18,  1.82s/it]
 81%|████████  | 11878/14725 [6:02:31<1:26:05,  1.81s/it]
 81%|████████  | 11879/14725 [6:02:33<1:27:09,  1.84s/it]
 81%|████████  | 11880/14725 [6:02:34<1:27:19,  1.84s/it]
                                                         

 81%|████████  | 11880/14725 [6:02:34<1:27:19,  1.84s/it]
 81%|████████  | 11881/14725 [6:02:36<1:26:28,  1.82s/it]
 81%|████████  | 11882/14725 [6:02:38<1:25:32,  1.81s/it]
 81%|████████  | 11883/14725 [6:02:40<1:25:43,  1.81s/it]
 81%|████████  | 11884/14725 [6
+0: {'loss': 0.8157, 'grad_norm': 0.7160867938793556, 'learning_rate': 2e-05, 'epoch': 0.81}
+0: :02:42<1:26:33,  1.83s/it]
 81%|████████  | 11885/14725 [6:02:43<1:27:09,  1.84s/it]
 81%|████████  | 11886/14725 [6:02:45<1:26:20,  1.82s/it]
 81%|████████  | 11887/14725 [6:02:47<1:26:15,  1.82s/it]
 81%|████████  | 11888/14725 [6:02:49<1:26:05,  1.82s/it]
 81%|████████  | 11889/14725 [6:02:51<1:26:12,  1.82s/it]
 81%|████████  | 11890/14725 [6:02:53<1:26:49,  1.84s/it]
                                                         

 81%|████████  | 11890/14725 [6:02:53<1:26:49,  1.84s/it]
 81%|████████  | 11891/14725 [6:02:54<1:26:13,  1.83s/it]
 81%|████████  | 11892/14725 [6:02:56<1:24:54,  1.80s/it]
 81%|████████  | 11893/14725 [6:02:58<1:25:32,  1.81s/it]
 81%|████████  | 11894/14725 [6:03:00<1:25:40,  1.82s/it]
 81%|████████  | 11895/14725 [6:03:02<1:24:47,  1.80s/it]
 81%|████████  | 11896/14725 [6:03
+0: {'loss': 0.8014, 'grad_norm': 0.6800497892098112, 'learning_rate': 2e-05, 'epoch': 0.81}
+0: :03<1:24:51,  1.80s/it]
 81%|████████  | 11897/14725 [6:03:05<1:24:37,  1.80s/it]
 81%|████████  | 11898/14725 [6:03:07<1:24:36,  1.80s/it]
 81%|████████  | 11899/14725 [6:03:09<1:25:12,  1.81s/it]
 81%|████████  | 11900/14725 [6:03:11<1:24:55,  1.80s/it]
                                                         

 81%|████████  | 11900/14725 [6:03:11<1:24:55,  1.80s/it]
 81%|████████  | 11901/14725 [6:03:12<1:25:35,  1.82s/it]
 81%|████████  | 11902/14725 [6:03:14<1:25:10,  1.81s/it]
 81%|████████  | 11903/14725 [6:03:16<1:25:54,  1.83s/it]
 81%|████████  | 11904/14725 [6:03:18<1:24:58,  1.81s/it]
 81%|████████  | 11905/14725 [6:03:20<1:25:14,  1.81s/it]
 81%|████████  | 11906/14725 [6:03:21<1:24:29,  1.80s/it]
 81%|████████  | 11907/14725 [6:03:23<1:24:19,  1.80s/it]
 81%|████████  | 11908/14725 [6:03:25
+0: {'loss': 0.7943, 'grad_norm': 0.7023164760025377, 'learning_rate': 2e-05, 'epoch': 0.81}
+0: <1:24:40,  1.80s/it]
 81%|████████  | 11909/14725 [6:03:27<1:25:10,  1.81s/it]
 81%|████████  | 11910/14725 [6:03:29<1:25:08,  1.81s/it]
                                                         

 81%|████████  | 11910/14725 [6:03:29<1:25:08,  1.81s/it]
 81%|████████  | 11911/14725 [6:03:31<1:24:58,  1.81s/it]
 81%|████████  | 11912/14725 [6:03:32<1:24:58,  1.81s/it]
 81%|████████  | 11913/14725 [6:03:34<1:24:34,  1.80s/it]
 81%|████████  | 11914/14725 [6:03:36<1:24:05,  1.79s/it]
 81%|████████  | 11915/14725 [6:03:38<1:24:19,  1.80s/it]
 81%|████████  | 11916/14725 [6:03:40<1:24:18,  1.80s/it]
 81%|████████  | 11917/14725 [6:03:41<1:24:00,  1.79s/it]
 81%|████████  | 11918/14725 [6:03:43<1:24:16,  1.80s/it]
 81%|████████  | 11919/14725 [6:03:45<1:24:21,  1.80s/it]
 81%|████████  | 11920/14725 [6:03:47<1:
+0: {'loss': 0.8173, 'grad_norm': 0.666313951093859, 'learning_rate': 2e-05, 'epoch': 0.81}
+0: {'loss': 0.7797, 'grad_norm': 0.6411125057749043, 'learning_rate': 2e-05, 'epoch': 0.81}
+0: 24:32,  1.81s/it]
                                                         

 81%|████████  | 11920/14725 [6:03:47<1:24:32,  1.81s/it]
 81%|████████  | 11921/14725 [6:03:49<1:24:32,  1.81s/it]
 81%|████████  | 11922/14725 [6:03:50<1:25:26,  1.83s/it]
 81%|████████  | 11923/14725 [6:03:52<1:24:22,  1.81s/it]
 81%|████████  | 11924/14725 [6:03:54<1:23:26,  1.79s/it]
 81%|████████  | 11925/14725 [6:03:56<1:23:16,  1.78s/it]
 81%|████████  | 11926/14725 [6:03:57<1:23:09,  1.78s/it]
 81%|████████  | 11927/14725 [6:03:59<1:24:23,  1.81s/it]
 81%|████████  | 11928/14725 [6:04:01<1:25:05,  1.83s/it]
 81%|████████  | 11929/14725 [6:04:03<1:25:10,  1.83s/it]
 81%|████████  | 11930/14725 [6:04:05<1:25:17,  1.83s/it]
                                                         

 81%|████████  | 11930/14725 [6:04:05<1:25:17,  1.83s/it]
+0: {'loss': 0.8154, 'grad_norm': 0.6604136610695162, 'learning_rate': 2e-05, 'epoch': 0.81}
+0:  81%|████████  | 11931/14725 [6:04:07<1:24:51,  1.82s/it]
 81%|████████  | 11932/14725 [6:04:08<1:24:40,  1.82s/it]
 81%|████████  | 11933/14725 [6:04:10<1:23:55,  1.80s/it]
 81%|████████  | 11934/14725 [6:04:12<1:24:14,  1.81s/it]
 81%|████████  | 11935/14725 [6:04:14<1:24:17,  1.81s/it]
 81%|████████  | 11936/14725 [6:04:16<1:23:37,  1.80s/it]
 81%|████████  | 11937/14725 [6:04:18<1:24:19,  1.81s/it]
 81%|████████  | 11938/14725 [6:04:19<1:23:52,  1.81s/it]
 81%|████████  | 11939/14725 [6:04:21<1:23:53,  1.81s/it]
 81%|████████  | 11940/14725 [6:04:23<1:23:38,  1.80s/it]
                                                         

 81%|████████  | 11940/14725 [6:04:23<1:23:38,  1.80s/it]
 81%|████████  | 11941/14725 [6:04:25<1:23:38,  1.80s/it]
 81%|████████  | 11942/14725 [6:04:27<1:23:40,  1.80s/it]
 81
+0: {'loss': 0.8085, 'grad_norm': 0.7042432050415244, 'learning_rate': 2e-05, 'epoch': 0.81}
+0: %|████████  | 11943/14725 [6:04:28<1:24:32,  1.82s/it]
 81%|████████  | 11944/14725 [6:04:30<1:24:15,  1.82s/it]
 81%|████████  | 11945/14725 [6:04:32<1:23:47,  1.81s/it]
 81%|████████  | 11946/14725 [6:04:34<1:23:05,  1.79s/it]
 81%|████████  | 11947/14725 [6:04:35<1:22:32,  1.78s/it]
 81%|████████  | 11948/14725 [6:04:37<1:21:48,  1.77s/it]
 81%|████████  | 11949/14725 [6:04:39<1:22:14,  1.78s/it]
 81%|████████  | 11950/14725 [6:04:41<1:22:15,  1.78s/it]
                                                         

 81%|████████  | 11950/14725 [6:04:41<1:22:15,  1.78s/it]
 81%|████████  | 11951/14725 [6:04:43<1:22:21,  1.78s/it]
 81%|████████  | 11952/14725 [6:04:44<1:22:14,  1.78s/it]
 81%|████████  | 11953/14725 [6:04:46<1:22:47,  1.79s/it]
 81%|████████  | 11954/14725 [6:04:48<1:23:40,  1.81s/it]
 81%|�
+0: {'loss': 0.7839, 'grad_norm': 0.6605755521423013, 'learning_rate': 2e-05, 'epoch': 0.81}
+0: ��███████  | 11955/14725 [6:04:50<1:23:59,  1.82s/it]
 81%|████████  | 11956/14725 [6:04:52<1:23:21,  1.81s/it]
 81%|████████  | 11957/14725 [6:04:53<1:22:49,  1.80s/it]
 81%|████████  | 11958/14725 [6:04:55<1:23:06,  1.80s/it]
 81%|████████  | 11959/14725 [6:04:57<1:23:22,  1.81s/it]
 81%|████████  | 11960/14725 [6:04:59<1:23:26,  1.81s/it]
                                                         

 81%|████████  | 11960/14725 [6:04:59<1:23:26,  1.81s/it]
 81%|████████  | 11961/14725 [6:05:01<1:23:14,  1.81s/it]
 81%|████████  | 11962/14725 [6:05:02<1:22:20,  1.79s/it]
 81%|████████  | 11963/14725 [6:05:04<1:22:03,  1.78s/it]
 81%|████████  | 11964/14725 [6:05:06<1:23:08,  1.81s/it]
 81%|████████▏ | 11965/14725 [6:05:08<1:22:31,  1.79s/it]
 81%|████████▏ | 11966/14725 [6:05:10<1:22:54,  1.80s/it]
 81%|
+0: {'loss': 0.7993, 'grad_norm': 0.6413423262679018, 'learning_rate': 2e-05, 'epoch': 0.81}
+0: ████████▏ | 11967/14725 [6:05:11<1:22:46,  1.80s/it]
 81%|████████▏ | 11968/14725 [6:05:13<1:22:10,  1.79s/it]
 81%|████████▏ | 11969/14725 [6:05:15<1:22:11,  1.79s/it]
 81%|████████▏ | 11970/14725 [6:05:17<1:21:36,  1.78s/it]
                                                         

 81%|████████▏ | 11970/14725 [6:05:17<1:21:36,  1.78s/it]
 81%|████████▏ | 11971/14725 [6:05:19<1:21:33,  1.78s/it]
 81%|████████▏ | 11972/14725 [6:05:20<1:21:54,  1.79s/it]
 81%|████████▏ | 11973/14725 [6:05:22<1:22:04,  1.79s/it]
 81%|████████▏ | 11974/14725 [6:05:24<1:22:36,  1.80s/it]
 81%|████████▏ | 11975/14725 [6:05:26<1:22:33,  1.80s/it]
 81%|████████▏ | 11976/14725 [6:05:28<1:22:26,  1.80s/it]
 81%|████████▏ | 11977/14725 [6:05:29<1:23:30,  1.82s/it]
 81%|████████▏ | 11978/14725 [6:05:31<1:
+0: {'loss': 0.8076, 'grad_norm': 0.6886429485033264, 'learning_rate': 2e-05, 'epoch': 0.81}
+0: 23:13,  1.82s/it]
 81%|████████▏ | 11979/14725 [6:05:33<1:23:07,  1.82s/it]
 81%|████████▏ | 11980/14725 [6:05:35<1:23:18,  1.82s/it]
                                                         

 81%|████████▏ | 11980/14725 [6:05:35<1:23:18,  1.82s/it]
 81%|████████▏ | 11981/14725 [6:05:37<1:23:25,  1.82s/it]
 81%|████████▏ | 11982/14725 [6:05:39<1:23:13,  1.82s/it]
 81%|████████▏ | 11983/14725 [6:05:40<1:23:26,  1.83s/it]
 81%|████████▏ | 11984/14725 [6:05:42<1:23:30,  1.83s/it]
 81%|████████▏ | 11985/14725 [6:05:44<1:23:13,  1.82s/it]
 81%|████████▏ | 11986/14725 [6:05:46<1:22:22,  1.80s/it]
 81%|████████▏ | 11987/14725 [6:05:48<1:22:56,  1.82s/it]
 81%|████████▏ | 11988/14725 [6:05:49<1:22:42,  1.81s/it]
 81%|████████▏ | 11989/14725 [6:05:51<1:22:01,  1.80s/it]
 81%|████████▏ | 
+0: {'loss': 0.7922, 'grad_norm': 0.6861844869372228, 'learning_rate': 2e-05, 'epoch': 0.81}
+0: {'loss': 0.7966, 'grad_norm': 0.7089574021298428, 'learning_rate': 2e-05, 'epoch': 0.81}
+0: 11990/14725 [6:05:53<1:21:33,  1.79s/it]
                                                         

 81%|████████▏ | 11990/14725 [6:05:53<1:21:33,  1.79s/it]
 81%|████████▏ | 11991/14725 [6:05:55<1:22:26,  1.81s/it]
 81%|████████▏ | 11992/14725 [6:05:57<1:22:02,  1.80s/it]
 81%|████████▏ | 11993/14725 [6:05:58<1:22:28,  1.81s/it]
 81%|████████▏ | 11994/14725 [6:06:00<1:22:08,  1.80s/it]
 81%|████████▏ | 11995/14725 [6:06:02<1:22:03,  1.80s/it]
 81%|████████▏ | 11996/14725 [6:06:04<1:22:57,  1.82s/it]
 81%|████████▏ | 11997/14725 [6:06:06<1:22:08,  1.81s/it]
 81%|████████▏ | 11998/14725 [6:06:07<1:22:18,  1.81s/it]
 81%|████████▏ | 11999/14725 [6:06:09<1:21:50,  1.80s/it]
 81%|████████▏ | 12000/14725 [6:06:11<1:21:46,  1.80s/it]
                                                         

 81%|████████
+0: {'loss': 0.8027, 'grad_norm': 0.659584216855264, 'learning_rate': 2e-05, 'epoch': 0.82}
+0: ▏ | 12000/14725 [6:06:11<1:21:46,  1.80s/it]
 82%|████████▏ | 12001/14725 [6:06:13<1:25:10,  1.88s/it]
 82%|████████▏ | 12002/14725 [6:06:15<1:24:22,  1.86s/it]
 82%|████████▏ | 12003/14725 [6:06:17<1:23:22,  1.84s/it]
 82%|████████▏ | 12004/14725 [6:06:19<1:22:41,  1.82s/it]
 82%|████████▏ | 12005/14725 [6:06:20<1:23:09,  1.83s/it]
 82%|████████▏ | 12006/14725 [6:06:22<1:22:44,  1.83s/it]
 82%|████████▏ | 12007/14725 [6:06:24<1:23:15,  1.84s/it]
 82%|████████▏ | 12008/14725 [6:06:26<1:23:34,  1.85s/it]
 82%|████████▏ | 12009/14725 [6:06:28<1:22:26,  1.82s/it]
 82%|████████▏ | 12010/14725 [6:06:29<1:22:26,  1.82s/it]
                                                         

 82%|████████▏ | 12010/14725 [6:06:29<1:22:26,  1.82s/it]
 82%|████████▏ | 12011/14725 [6:06:31<1:22:14,  1.82s/it]
 82%|�
+0: {'loss': 0.7812, 'grad_norm': 0.6687151499932811, 'learning_rate': 2e-05, 'epoch': 0.82}
+0: ��███████▏ | 12012/14725 [6:06:33<1:21:48,  1.81s/it]
 82%|████████▏ | 12013/14725 [6:06:35<1:22:00,  1.81s/it]
 82%|████████▏ | 12014/14725 [6:06:37<1:21:36,  1.81s/it]
 82%|████████▏ | 12015/14725 [6:06:38<1:20:33,  1.78s/it]
 82%|████████▏ | 12016/14725 [6:06:40<1:20:38,  1.79s/it]
 82%|████████▏ | 12017/14725 [6:06:42<1:20:03,  1.77s/it]
 82%|████████▏ | 12018/14725 [6:06:44<1:20:39,  1.79s/it]
 82%|████████▏ | 12019/14725 [6:06:46<1:21:55,  1.82s/it]
 82%|████████▏ | 12020/14725 [6:06:47<1:21:23,  1.81s/it]
                                                         

 82%|████████▏ | 12020/14725 [6:06:47<1:21:23,  1.81s/it]
 82%|████████▏ | 12021/14725 [6:06:49<1:22:53,  1.84s/it]
 82%|████████▏ | 12022/14725 [6:06:51<1:22:35,  1.83s/it]
 82%|████████▏ | 12023/14725 [6:06:53<1:2
+0: {'loss': 0.7918, 'grad_norm': 0.6890431599775999, 'learning_rate': 2e-05, 'epoch': 0.82}
+0: 1:29,  1.81s/it]
 82%|████████▏ | 12024/14725 [6:06:55<1:21:10,  1.80s/it]
 82%|████████▏ | 12025/14725 [6:06:57<1:21:49,  1.82s/it]
 82%|████████▏ | 12026/14725 [6:06:58<1:21:42,  1.82s/it]
 82%|████████▏ | 12027/14725 [6:07:00<1:21:30,  1.81s/it]
 82%|████████▏ | 12028/14725 [6:07:02<1:22:11,  1.83s/it]
 82%|████████▏ | 12029/14725 [6:07:04<1:22:12,  1.83s/it]
 82%|████████▏ | 12030/14725 [6:07:06<1:22:56,  1.85s/it]
                                                         

 82%|████████▏ | 12030/14725 [6:07:06<1:22:56,  1.85s/it]
 82%|████████▏ | 12031/14725 [6:07:08<1:23:35,  1.86s/it]
 82%|████████▏ | 12032/14725 [6:07:10<1:23:19,  1.86s/it]
 82%|████████▏ | 12033/14725 [6:07:11<1:22:45,  1.84s/it]
 82%|████████▏ | 12034/14725 [6:07:13<1:21:56,  1.83s/it]
 82%|████████▏ | 1
+0: {'loss': 0.817, 'grad_norm': 0.6843015162475946, 'learning_rate': 2e-05, 'epoch': 0.82}
+0: 2035/14725 [6:07:15<1:22:06,  1.83s/it]
 82%|████████▏ | 12036/14725 [6:07:17<1:22:44,  1.85s/it]
 82%|████████▏ | 12037/14725 [6:07:19<1:22:43,  1.85s/it]
 82%|████████▏ | 12038/14725 [6:07:20<1:21:53,  1.83s/it]
 82%|████████▏ | 12039/14725 [6:07:22<1:21:56,  1.83s/it]
 82%|████████▏ | 12040/14725 [6:07:24<1:20:34,  1.80s/it]
                                                         

 82%|████████▏ | 12040/14725 [6:07:24<1:20:34,  1.80s/it]
 82%|████████▏ | 12041/14725 [6:07:26<1:21:05,  1.81s/it]
 82%|████████▏ | 12042/14725 [6:07:28<1:20:56,  1.81s/it]
 82%|████████▏ | 12043/14725 [6:07:29<1:20:49,  1.81s/it]
 82%|████████▏ | 12044/14725 [6:07:31<1:21:23,  1.82s/it]
 82%|████████▏ | 12045/14725 [6:07:33<1:21:03,  1.81s/it]
 82%|████████▏ | 12046/14725 [6:07:35<1:20:38,  1.81s/it]
 82%|██�
+0: {'loss': 0.817, 'grad_norm': 0.6613603638436275, 'learning_rate': 2e-05, 'epoch': 0.82}
+0: �█████▏ | 12047/14725 [6:07:37<1:20:16,  1.80s/it]
 82%|████████▏ | 12048/14725 [6:07:38<1:19:58,  1.79s/it]
 82%|████████▏ | 12049/14725 [6:07:40<1:20:32,  1.81s/it]
 82%|████████▏ | 12050/14725 [6:07:42<1:20:02,  1.80s/it]
                                                         

 82%|████████▏ | 12050/14725 [6:07:42<1:20:02,  1.80s/it]
 82%|████████▏ | 12051/14725 [6:07:44<1:20:20,  1.80s/it]
 82%|████████▏ | 12052/14725 [6:07:46<1:19:57,  1.79s/it]
 82%|████████▏ | 12053/14725 [6:07:48<1:20:32,  1.81s/it]
 82%|████████▏ | 12054/14725 [6:07:49<1:20:44,  1.81s/it]
 82%|████████▏ | 12055/14725 [6:07:51<1:20:51,  1.82s/it]
 82%|████████▏ | 12056/14725 [6:07:53<1:20:49,  1.82s/it]
 82%|████████▏ | 12057/14725 [6:07:55<1:20:38,  1.81s/it]
 82%|████████▏ | 12058/14725 [6:07:57<1:20:55,  
+0: {'loss': 0.797, 'grad_norm': 0.6486000614668483, 'learning_rate': 2e-05, 'epoch': 0.82}
+0: 1.82s/it]
 82%|████████▏ | 12059/14725 [6:07:58<1:21:19,  1.83s/it]
 82%|████████▏ | 12060/14725 [6:08:00<1:20:54,  1.82s/it]
                                                         

 82%|████████▏ | 12060/14725 [6:08:00<1:20:54,  1.82s/it]
 82%|████████▏ | 12061/14725 [6:08:02<1:20:29,  1.81s/it]
 82%|████████▏ | 12062/14725 [6:08:04<1:20:06,  1.81s/it]
 82%|████████▏ | 12063/14725 [6:08:06<1:20:44,  1.82s/it]
 82%|████████▏ | 12064/14725 [6:08:08<1:20:15,  1.81s/it]
 82%|████████▏ | 12065/14725 [6:08:09<1:19:49,  1.80s/it]
 82%|████████▏ | 12066/14725 [6:08:11<1:19:30,  1.79s/it]
 82%|████████▏ | 12067/14725 [6:08:13<1:18:57,  1.78s/it]
 82%|████████▏ | 12068/14725 [6:08:15<1:19:03,  1.79s/it]
 82%|████████▏ | 12069/14725 [6:08:16<1:18:56,  1.78s/it]
 82%|████████▏ | 12070/14
+0: {'loss': 0.8061, 'grad_norm': 0.6746261769061767, 'learning_rate': 2e-05, 'epoch': 0.82}
+0: {'loss': 0.8028, 'grad_norm': 0.6889695047918414, 'learning_rate': 2e-05, 'epoch': 0.82}
+0: 725 [6:08:18<1:18:54,  1.78s/it]
                                                         

 82%|████████▏ | 12070/14725 [6:08:18<1:18:54,  1.78s/it]
 82%|████████▏ | 12071/14725 [6:08:20<1:19:33,  1.80s/it]
 82%|████████▏ | 12072/14725 [6:08:22<1:19:20,  1.79s/it]
 82%|████████▏ | 12073/14725 [6:08:24<1:19:01,  1.79s/it]
 82%|████████▏ | 12074/14725 [6:08:25<1:19:04,  1.79s/it]
 82%|████████▏ | 12075/14725 [6:08:27<1:20:05,  1.81s/it]
 82%|████████▏ | 12076/14725 [6:08:29<1:20:04,  1.81s/it]
 82%|████████▏ | 12077/14725 [6:08:31<1:19:34,  1.80s/it]
 82%|████████▏ | 12078/14725 [6:08:33<1:19:54,  1.81s/it]
 82%|████████▏ | 12079/14725 [6:08:34<1:20:03,  1.82s/it]
 82%|████████▏ | 12080/14725 [6:08:36<1:20:05,  1.82s/it]
                                                         

 82%|████████▏ | 12
+0: {'loss': 0.7907, 'grad_norm': 0.6699770067296927, 'learning_rate': 2e-05, 'epoch': 0.82}
+0: 080/14725 [6:08:36<1:20:05,  1.82s/it]
 82%|████████▏ | 12081/14725 [6:08:38<1:19:25,  1.80s/it]
 82%|████████▏ | 12082/14725 [6:08:40<1:19:39,  1.81s/it]
 82%|████████▏ | 12083/14725 [6:08:42<1:20:17,  1.82s/it]
 82%|████████▏ | 12084/14725 [6:08:44<1:21:02,  1.84s/it]
 82%|████████▏ | 12085/14725 [6:08:45<1:20:26,  1.83s/it]
 82%|████████▏ | 12086/14725 [6:08:47<1:20:10,  1.82s/it]
 82%|████████▏ | 12087/14725 [6:08:49<1:19:54,  1.82s/it]
 82%|████████▏ | 12088/14725 [6:08:51<1:19:40,  1.81s/it]
 82%|████████▏ | 12089/14725 [6:08:53<1:19:04,  1.80s/it]
 82%|████████▏ | 12090/14725 [6:08:54<1:18:35,  1.79s/it]
                                                         

 82%|████████▏ | 12090/14725 [6:08:54<1:18:35,  1.79s/it]
 82%|████████▏ | 12091/14725 [6:08:56<1:20:12,  1.83s/it]
 82%|███
+0: {'loss': 0.783, 'grad_norm': 0.6285207015043328, 'learning_rate': 2e-05, 'epoch': 0.82}
+0: █████▏ | 12092/14725 [6:08:58<1:19:42,  1.82s/it]
 82%|████████▏ | 12093/14725 [6:09:00<1:19:10,  1.80s/it]
 82%|████████▏ | 12094/14725 [6:09:02<1:19:28,  1.81s/it]
 82%|████████▏ | 12095/14725 [6:09:03<1:19:00,  1.80s/it]
 82%|████████▏ | 12096/14725 [6:09:05<1:18:44,  1.80s/it]
 82%|████████▏ | 12097/14725 [6:09:07<1:18:19,  1.79s/it]
 82%|████████▏ | 12098/14725 [6:09:09<1:18:26,  1.79s/it]
 82%|████████▏ | 12099/14725 [6:09:11<1:17:53,  1.78s/it]
 82%|████████▏ | 12100/14725 [6:09:13<1:20:58,  1.85s/it]
                                                         

 82%|████████▏ | 12100/14725 [6:09:13<1:20:58,  1.85s/it]
 82%|████████▏ | 12101/14725 [6:09:14<1:21:08,  1.86s/it]
 82%|████████▏ | 12102/14725 [6:09:16<1:19:49,  1.83s/it]
 82%|████████▏ | 12103/14725 [6:09:18<1:20:01,  1
+0: {'loss': 0.8247, 'grad_norm': 0.6897918161040234, 'learning_rate': 2e-05, 'epoch': 0.82}
+0: .83s/it]
 82%|████████▏ | 12104/14725 [6:09:20<1:20:06,  1.83s/it]
 82%|████████▏ | 12105/14725 [6:09:22<1:19:00,  1.81s/it]
 82%|████████▏ | 12106/14725 [6:09:23<1:19:02,  1.81s/it]
 82%|████████▏ | 12107/14725 [6:09:25<1:19:02,  1.81s/it]
 82%|████████▏ | 12108/14725 [6:09:27<1:18:26,  1.80s/it]
 82%|████████▏ | 12109/14725 [6:09:29<1:19:11,  1.82s/it]
 82%|████████▏ | 12110/14725 [6:09:31<1:19:06,  1.82s/it]
                                                         

 82%|████████▏ | 12110/14725 [6:09:31<1:19:06,  1.82s/it]
 82%|████████▏ | 12111/14725 [6:09:33<1:19:01,  1.81s/it]
 82%|████████▏ | 12112/14725 [6:09:34<1:18:31,  1.80s/it]
 82%|████████▏ | 12113/14725 [6:09:36<1:18:22,  1.80s/it]
 82%|████████▏ | 12114/14725 [6:09:38<1:19:31,  1.83s/it]
 82%|████████▏ | 12115/147
+0: {'loss': 0.7828, 'grad_norm': 0.670335283793295, 'learning_rate': 2e-05, 'epoch': 0.82}
+0: 25 [6:09:40<1:19:05,  1.82s/it]
 82%|████████▏ | 12116/14725 [6:09:42<1:18:49,  1.81s/it]
 82%|████████▏ | 12117/14725 [6:09:43<1:17:53,  1.79s/it]
 82%|████████▏ | 12118/14725 [6:09:45<1:18:20,  1.80s/it]
 82%|████████▏ | 12119/14725 [6:09:47<1:18:52,  1.82s/it]
 82%|████████▏ | 12120/14725 [6:09:49<1:18:15,  1.80s/it]
                                                         

 82%|████████▏ | 12120/14725 [6:09:49<1:18:15,  1.80s/it]
 82%|████████▏ | 12121/14725 [6:09:51<1:18:31,  1.81s/it]
 82%|████████▏ | 12122/14725 [6:09:52<1:18:16,  1.80s/it]
 82%|████████▏ | 12123/14725 [6:09:54<1:19:53,  1.84s/it]
 82%|████████▏ | 12124/14725 [6:09:56<1:18:47,  1.82s/it]
 82%|████████▏ | 12125/14725 [6:09:58<1:18:21,  1.81s/it]
 82%|████████▏ | 12126/14725 [6:10:00<1:18:32,  1.81s/it]
 82%|█████�
+0: {'loss': 0.8169, 'grad_norm': 0.6620662902588358, 'learning_rate': 2e-05, 'epoch': 0.82}
+0: ��██▏ | 12127/14725 [6:10:02<1:18:35,  1.82s/it]
 82%|████████▏ | 12128/14725 [6:10:03<1:20:04,  1.85s/it]
 82%|████████▏ | 12129/14725 [6:10:05<1:19:23,  1.83s/it]
 82%|████████▏ | 12130/14725 [6:10:07<1:18:56,  1.83s/it]
                                                         

 82%|████████▏ | 12130/14725 [6:10:07<1:18:56,  1.83s/it]
 82%|████████▏ | 12131/14725 [6:10:09<1:19:41,  1.84s/it]
 82%|████████▏ | 12132/14725 [6:10:11<1:18:16,  1.81s/it]
 82%|████████▏ | 12133/14725 [6:10:12<1:17:55,  1.80s/it]
 82%|████████▏ | 12134/14725 [6:10:14<1:18:36,  1.82s/it]
 82%|████████▏ | 12135/14725 [6:10:16<1:19:01,  1.83s/it]
 82%|████████▏ | 12136/14725 [6:10:18<1:18:06,  1.81s/it]
 82%|████████▏ | 12137/14725 [6:10:20<1:18:01,  1.81s/it]
 82%|████████▏ | 12138/14725 [6:10:22<1:17:47,  1.80s/it
+0: {'loss': 0.7801, 'grad_norm': 0.6662287629602711, 'learning_rate': 2e-05, 'epoch': 0.82}
+0: ]
 82%|████████▏ | 12139/14725 [6:10:23<1:18:04,  1.81s/it]
 82%|████████▏ | 12140/14725 [6:10:25<1:19:16,  1.84s/it]
                                                         

 82%|████████▏ | 12140/14725 [6:10:25<1:19:16,  1.84s/it]
 82%|████████▏ | 12141/14725 [6:10:27<1:18:19,  1.82s/it]
 82%|████████▏ | 12142/14725 [6:10:29<1:18:00,  1.81s/it]
 82%|████████▏ | 12143/14725 [6:10:31<1:17:59,  1.81s/it]
 82%|████████▏ | 12144/14725 [6:10:32<1:17:28,  1.80s/it]
 82%|████████▏ | 12145/14725 [6:10:34<1:18:33,  1.83s/it]
 82%|████████▏ | 12146/14725 [6:10:36<1:18:04,  1.82s/it]
 82%|████████▏ | 12147/14725 [6:10:38<1:17:00,  1.79s/it]
 82%|████████▏ | 12148/14725 [6:10:40<1:17:03,  1.79s/it]
 83%|████████▎ | 12149/14725 [6:10:41<1:17:26,  1.80s/it]
 83%|████████▎ | 12150/14725 [6:1
+0: {'loss': 0.7762, 'grad_norm': 0.7101121670555692, 'learning_rate': 2e-05, 'epoch': 0.83}
+0: {'loss': 0.7993, 'grad_norm': 0.7619644186182741, 'learning_rate': 2e-05, 'epoch': 0.83}
+0: 0:43<1:18:17,  1.82s/it]
                                                         

 83%|████████▎ | 12150/14725 [6:10:43<1:18:17,  1.82s/it]
 83%|████████▎ | 12151/14725 [6:10:45<1:17:46,  1.81s/it]
 83%|████████▎ | 12152/14725 [6:10:47<1:17:25,  1.81s/it]
 83%|████████▎ | 12153/14725 [6:10:49<1:16:50,  1.79s/it]
 83%|████████▎ | 12154/14725 [6:10:51<1:17:39,  1.81s/it]
 83%|████████▎ | 12155/14725 [6:10:52<1:17:36,  1.81s/it]
 83%|████████▎ | 12156/14725 [6:10:54<1:17:02,  1.80s/it]
 83%|████████▎ | 12157/14725 [6:10:56<1:16:19,  1.78s/it]
 83%|████████▎ | 12158/14725 [6:10:58<1:16:17,  1.78s/it]
 83%|████████▎ | 12159/14725 [6:10:59<1:16:11,  1.78s/it]
 83%|████████▎ | 12160/14725 [6:11:02<1:22:21,  1.93s/it]
                                                         

 83%|████████▎ | 12160/1472
+0: {'loss': 0.8074, 'grad_norm': 0.6685518898981359, 'learning_rate': 2e-05, 'epoch': 0.83}
+0: 5 [6:11:02<1:22:21,  1.93s/it]
 83%|████████▎ | 12161/14725 [6:11:04<1:21:25,  1.91s/it]
 83%|████████▎ | 12162/14725 [6:11:05<1:20:16,  1.88s/it]
 83%|████████▎ | 12163/14725 [6:11:07<1:20:20,  1.88s/it]
 83%|████████▎ | 12164/14725 [6:11:09<1:19:42,  1.87s/it]
 83%|████████▎ | 12165/14725 [6:11:11<1:19:47,  1.87s/it]
 83%|████████▎ | 12166/14725 [6:11:13<1:19:23,  1.86s/it]
 83%|████��███▎ | 12167/14725 [6:11:15<1:18:28,  1.84s/it]
 83%|████████▎ | 12168/14725 [6:11:16<1:17:56,  1.83s/it]
 83%|████████▎ | 12169/14725 [6:11:18<1:18:35,  1.85s/it]
 83%|████████▎ | 12170/14725 [6:11:20<1:18:04,  1.83s/it]
                                                         

 83%|████████▎ | 12170/14725 [6:11:20<1:18:04,  1.83s/it]
 83%|████████▎ | 12171/14725 [6:11:22<1:17:27,  1.82s/it]
 83%|█████�
+0: {'loss': 0.8043, 'grad_norm': 0.6698511952728812, 'learning_rate': 2e-05, 'epoch': 0.83}
+0: �██▎ | 12172/14725 [6:11:24<1:16:31,  1.80s/it]
 83%|████████▎ | 12173/14725 [6:11:25<1:16:19,  1.79s/it]
 83%|████████▎ | 12174/14725 [6:11:27<1:16:55,  1.81s/it]
 83%|████████▎ | 12175/14725 [6:11:29<1:16:06,  1.79s/it]
 83%|████████▎ | 12176/14725 [6:11:31<1:20:58,  1.91s/it]
 83%|████████▎ | 12177/14725 [6:11:33<1:19:46,  1.88s/it]
 83%|████████▎ | 12178/14725 [6:11:35<1:19:18,  1.87s/it]
 83%|████████▎ | 12179/14725 [6:11:37<1:18:51,  1.86s/it]
 83%|████████▎ | 12180/14725 [6:11:38<1:17:48,  1.83s/it]
                                                         

 83%|████████▎ | 12180/14725 [6:11:38<1:17:48,  1.83s/it]
 83%|████████▎ | 12181/14725 [6:11:40<1:19:27,  1.87s/it]
 83%|████████▎ | 12182/14725 [6:11:42<1:17:46,  1.84s/it]
 83%|████████▎ | 12183/14725 [6:11:44<1:16:47,  1.81s/it]
+0: {'loss': 0.8008, 'grad_norm': 0.7147157364667418, 'learning_rate': 2e-05, 'epoch': 0.83}
+0: 
 83%|████████▎ | 12184/14725 [6:11:46<1:16:26,  1.81s/it]
 83%|████████▎ | 12185/14725 [6:11:47<1:15:38,  1.79s/it]
 83%|████████▎ | 12186/14725 [6:11:49<1:15:12,  1.78s/it]
 83%|████████▎ | 12187/14725 [6:11:51<1:15:40,  1.79s/it]
 83%|████████▎ | 12188/14725 [6:11:53<1:15:23,  1.78s/it]
 83%|████████▎ | 12189/14725 [6:11:55<1:16:02,  1.80s/it]
 83%|████████▎ | 12190/14725 [6:11:56<1:15:53,  1.80s/it]
                                                         

 83%|████████▎ | 12190/14725 [6:11:56<1:15:53,  1.80s/it]
 83%|████████▎ | 12191/14725 [6:11:58<1:15:29,  1.79s/it]
 83%|████████▎ | 12192/14725 [6:12:00<1:15:22,  1.79s/it]
 83%|████████▎ | 12193/14725 [6:12:02<1:14:50,  1.77s/it]
 83%|████████▎ | 12194/14725 [6:12:04<1:15:14,  1.78s/it]
 83%|████████▎ | 12195/14725 [6:12
+0: {'loss': 0.8143, 'grad_norm': 0.6492007120924046, 'learning_rate': 2e-05, 'epoch': 0.83}
+0: :05<1:15:48,  1.80s/it]
 83%|████████▎ | 12196/14725 [6:12:07<1:20:09,  1.90s/it]
 83%|████████▎ | 12197/14725 [6:12:09<1:18:59,  1.87s/it]
 83%|████████▎ | 12198/14725 [6:12:11<1:17:16,  1.83s/it]
 83%|████████▎ | 12199/14725 [6:12:13<1:17:20,  1.84s/it]
 83%|████████▎ | 12200/14725 [6:12:15<1:16:01,  1.81s/it]
                                                         

 83%|████████▎ | 12200/14725 [6:12:15<1:16:01,  1.81s/it]
 83%|████████▎ | 12201/14725 [6:12:16<1:15:43,  1.80s/it]
 83%|████████▎ | 12202/14725 [6:12:18<1:16:54,  1.83s/it]
 83%|████████▎ | 12203/14725 [6:12:20<1:16:49,  1.83s/it]
 83%|████████▎ | 12204/14725 [6:12:22<1:17:37,  1.85s/it]
 83%|████████▎ | 12205/14725 [6:12:24<1:16:57,  1.83s/it]
 83%|████████▎ | 12206/14725 [6:12:26<1:16:21,  1.82s/it]
 83%|████████
+0: {'loss': 0.7874, 'grad_norm': 0.6771447830153844, 'learning_rate': 2e-05, 'epoch': 0.83}
+0: ▎ | 12207/14725 [6:12:27<1:15:51,  1.81s/it]
 83%|████████▎ | 12208/14725 [6:12:29<1:16:20,  1.82s/it]
 83%|████████▎ | 12209/14725 [6:12:31<1:16:44,  1.83s/it]
 83%|████████▎ | 12210/14725 [6:12:33<1:15:56,  1.81s/it]
                                                         

 83%|████████▎ | 12210/14725 [6:12:33<1:15:56,  1.81s/it]
 83%|████████▎ | 12211/14725 [6:12:35<1:19:37,  1.90s/it]
 83%|████████▎ | 12212/14725 [6:12:37<1:18:42,  1.88s/it]
 83%|████████▎ | 12213/14725 [6:12:39<1:18:12,  1.87s/it]
 83%|████████▎ | 12214/14725 [6:12:41<1:18:32,  1.88s/it]
 83%|████████▎ | 12215/14725 [6:12:42<1:17:16,  1.85s/it]
 83%|████████▎ | 12216/14725 [6:12:44<1:16:45,  1.84s/it]
 83%|████████▎ | 12217/14725 [6:12:46<1:15:58,  1.82s/it]
 83%|████████▎ | 12218/14725 [6:12:48<1:15:00,  1.80s/it]
 83%|�
+0: {'loss': 0.7947, 'grad_norm': 0.6692513821236218, 'learning_rate': 2e-05, 'epoch': 0.83}
+0: ��████��██▎ | 12219/14725 [6:12:50<1:15:48,  1.81s/it]
 83%|████████▎ | 12220/14725 [6:12:51<1:16:30,  1.83s/it]
                                                         

 83%|████████▎ | 12220/14725 [6:12:51<1:16:30,  1.83s/it]
 83%|████████▎ | 12221/14725 [6:12:53<1:15:46,  1.82s/it]
 83%|████████▎ | 12222/14725 [6:12:55<1:15:10,  1.80s/it]
 83%|████████▎ | 12223/14725 [6:12:57<1:15:44,  1.82s/it]
 83%|████████▎ | 12224/14725 [6:12:59<1:15:56,  1.82s/it]
 83%|████████▎ | 12225/14725 [6:13:00<1:16:09,  1.83s/it]
 83%|████████▎ | 12226/14725 [6:13:02<1:15:16,  1.81s/it]
 83%|████████▎ | 12227/14725 [6:13:04<1:15:54,  1.82s/it]
 83%|████████▎ | 12228/14725 [6:13:06<1:15:05,  1.80s/it]
 83%|████████▎ | 12229/14725 [6:13:08<1:15:17,  1.81s/it]
 83%|████████▎ | 12230/14725 [6:13:09<1:1
+0: {'loss': 0.7912, 'grad_norm': 0.6727107884496337, 'learning_rate': 2e-05, 'epoch': 0.83}
+0: {'loss': 0.7908, 'grad_norm': 0.6771185704791445, 'learning_rate': 2e-05, 'epoch': 0.83}
+0: 4:31,  1.79s/it]
                                                         

 83%|████████▎ | 12230/14725 [6:13:09<1:14:31,  1.79s/it]
 83%|████████▎ | 12231/14725 [6:13:11<1:14:39,  1.80s/it]
 83%|████████▎ | 12232/14725 [6:13:13<1:14:22,  1.79s/it]
 83%|████████▎ | 12233/14725 [6:13:15<1:14:08,  1.79s/it]
 83%|████████▎ | 12234/14725 [6:13:17<1:15:10,  1.81s/it]
 83%|████████▎ | 12235/14725 [6:13:18<1:14:16,  1.79s/it]
 83%|████████▎ | 12236/14725 [6:13:20<1:14:46,  1.80s/it]
 83%|████████▎ | 12237/14725 [6:13:22<1:14:43,  1.80s/it]
 83%|████████▎ | 12238/14725 [6:13:24<1:14:06,  1.79s/it]
 83%|████████▎ | 12239/14725 [6:13:26<1:13:55,  1.78s/it]
 83%|████████▎ | 12240/14725 [6:13:27<1:14:17,  1.79s/it]
                                                         

 83%|████████▎ | 12240/14725 [6:13:
+0: {'loss': 0.7946, 'grad_norm': 0.661837627841428, 'learning_rate': 2e-05, 'epoch': 0.83}
+0: 27<1:14:17,  1.79s/it]
 83%|████████▎ | 12241/14725 [6:13:29<1:14:40,  1.80s/it]
 83%|████████▎ | 12242/14725 [6:13:31<1:13:58,  1.79s/it]
 83%|████████▎ | 12243/14725 [6:13:33<1:13:31,  1.78s/it]
 83%|████████▎ | 12244/14725 [6:13:34<1:13:50,  1.79s/it]
 83%|████████▎ | 12245/14725 [6:13:36<1:13:29,  1.78s/it]
 83%|████████▎ | 12246/14725 [6:13:38<1:13:18,  1.77s/it]
 83%|████████▎ | 12247/14725 [6:13:40<1:13:55,  1.79s/it]
 83%|████████▎ | 12248/14725 [6:13:42<1:14:04,  1.79s/it]
 83%|████████▎ | 12249/14725 [6:13:44<1:14:47,  1.81s/it]
 83%|████████▎ | 12250/14725 [6:13:45<1:14:30,  1.81s/it]
                                                         

 83%|████████▎ | 12250/14725 [6:13:45<1:14:30,  1.81s/it]
 83%|████████▎ | 12251/14725 [6:13:47<1:14:01,  1.80s/it]
 83%|████████�
+0: {'loss': 0.7793, 'grad_norm': 0.682162997158338, 'learning_rate': 2e-05, 'epoch': 0.83}
+0: �� | 12252/14725 [6:13:49<1:13:45,  1.79s/it]
 83%|████████▎ | 12253/14725 [6:13:51<1:13:47,  1.79s/it]
 83%|████████▎ | 12254/14725 [6:13:52<1:13:40,  1.79s/it]
 83%|████████▎ | 12255/14725 [6:13:54<1:13:44,  1.79s/it]
 83%|████████▎ | 12256/14725 [6:13:56<1:13:40,  1.79s/it]
 83%|████████▎ | 12257/14725 [6:13:58<1:13:56,  1.80s/it]
 83%|████████▎ | 12258/14725 [6:14:00<1:13:40,  1.79s/it]
 83%|████████▎ | 12259/14725 [6:14:01<1:13:14,  1.78s/it]
 83%|████████▎ | 12260/14725 [6:14:03<1:13:00,  1.78s/it]
                                                         

 83%|████████▎ | 12260/14725 [6:14:03<1:13:00,  1.78s/it]
 83%|████████▎ | 12261/14725 [6:14:05<1:14:21,  1.81s/it]
 83%|████████▎ | 12262/14725 [6:14:07<1:14:32,  1.82s/it]
 83%|████████▎ | 12263/14725 [6:14:09<1:14:04,  1.81s/it]
 83%|�
+0: {'loss': 0.7831, 'grad_norm': 0.6769093274745601, 'learning_rate': 2e-05, 'epoch': 0.83}
+0: �███████▎ | 12264/14725 [6:14:10<1:13:37,  1.79s/it]
 83%|████████▎ | 12265/14725 [6:14:12<1:14:29,  1.82s/it]
 83%|████████▎ | 12266/14725 [6:14:14<1:14:27,  1.82s/it]
 83%|████████▎ | 12267/14725 [6:14:16<1:14:00,  1.81s/it]
 83%|████████▎ | 12268/14725 [6:14:18<1:13:35,  1.80s/it]
 83%|████████▎ | 12269/14725 [6:14:19<1:13:50,  1.80s/it]
 83%|████████▎ | 12270/14725 [6:14:21<1:14:59,  1.83s/it]
                                                         

 83%|████████▎ | 12270/14725 [6:14:21<1:14:59,  1.83s/it]
 83%|████████▎ | 12271/14725 [6:14:23<1:14:53,  1.83s/it]
 83%|████████▎ | 12272/14725 [6:14:25<1:15:51,  1.86s/it]
 83%|████████▎ | 12273/14725 [6:14:27<1:14:38,  1.83s/it]
 83%|████████▎ | 12274/14725 [6:14:29<1:14:33,  1.83s/it]
 83%|████████▎ | 12275/14725 [6:14:30<1:14
+0: {'loss': 0.8072, 'grad_norm': 0.6566329707404042, 'learning_rate': 2e-05, 'epoch': 0.83}
+0: :08,  1.82s/it]
 83%|████████▎ | 12276/14725 [6:14:32<1:14:42,  1.83s/it]
 83%|████████▎ | 12277/14725 [6:14:34<1:13:58,  1.81s/it]
 83%|████████▎ | 12278/14725 [6:14:36<1:15:09,  1.84s/it]
 83%|████████▎ | 12279/14725 [6:14:38<1:14:48,  1.83s/it]
 83%|████████▎ | 12280/14725 [6:14:40<1:13:55,  1.81s/it]
                                                         

 83%|████████▎ | 12280/14725 [6:14:40<1:13:55,  1.81s/it]
 83%|████████▎ | 12281/14725 [6:14:41<1:13:18,  1.80s/it]
 83%|████████▎ | 12282/14725 [6:14:43<1:12:52,  1.79s/it]
 83%|████████▎ | 12283/14725 [6:14:45<1:14:21,  1.83s/it]
 83%|████████▎ | 12284/14725 [6:14:47<1:13:33,  1.81s/it]
 83%|████████▎ | 12285/14725 [6:14:49<1:13:14,  1.80s/it]
 83%|████████▎ | 12286/14725 [6:14:50<1:13:35,  1.81s/it]
 83%|████████▎ | 12
+0: {'loss': 0.8133, 'grad_norm': 0.647763173603013, 'learning_rate': 2e-05, 'epoch': 0.83}
+0: 287/14725 [6:14:52<1:12:56,  1.80s/it]
 83%|████████▎ | 12288/14725 [6:14:54<1:14:00,  1.82s/it]
 83%|████████▎ | 12289/14725 [6:14:56<1:13:13,  1.80s/it]
 83%|████████▎ | 12290/14725 [6:14:58<1:12:27,  1.79s/it]
                                                         

 83%|████████▎ | 12290/14725 [6:14:58<1:12:27,  1.79s/it]
 83%|████████▎ | 12291/14725 [6:14:59<1:12:12,  1.78s/it]
 83%|████████▎ | 12292/14725 [6:15:01<1:12:48,  1.80s/it]
 83%|████████▎ | 12293/14725 [6:15:03<1:11:59,  1.78s/it]
 83%|████████▎ | 12294/14725 [6:15:05<1:11:47,  1.77s/it]
 83%|████████▎ | 12295/14725 [6:15:07<1:12:46,  1.80s/it]
 84%|████████▎ | 12296/14725 [6:15:08<1:12:11,  1.78s/it]
 84%|████████▎ | 12297/14725 [6:15:10<1:14:00,  1.83s/it]
 84%|████████▎ | 12298/14725 [6:15:12<1:14:23,  1.84s/it]
 84%|███
+0: {'loss': 0.7928, 'grad_norm': 0.6023045239268325, 'learning_rate': 2e-05, 'epoch': 0.84}
+0: █████▎ | 12299/14725 [6:15:14<1:13:39,  1.82s/it]
 84%|████████▎ | 12300/14725 [6:15:16<1:13:00,  1.81s/it]
                                                         

 84%|████████▎ | 12300/14725 [6:15:16<1:13:00,  1.81s/it]
 84%|████████▎ | 12301/14725 [6:15:18<1:14:04,  1.83s/it]
 84%|████████▎ | 12302/14725 [6:15:19<1:13:51,  1.83s/it]
 84%|████████▎ | 12303/14725 [6:15:21<1:14:12,  1.84s/it]
 84%|████████▎ | 12304/14725 [6:15:23<1:13:32,  1.82s/it]
 84%|████████▎ | 12305/14725 [6:15:25<1:12:27,  1.80s/it]
 84%|████████▎ | 12306/14725 [6:15:27<1:13:33,  1.82s/it]
 84%|████████▎ | 12307/14725 [6:15:28<1:12:54,  1.81s/it]
 84%|████████▎ | 12308/14725 [6:15:30<1:11:56,  1.79s/it]
 84%|████████▎ | 12309/14725 [6:15:32<1:12:10,  1.79s/it]
 84%|████████▎ | 12310/14725 [6:15:34<1:13:27,  1
+0: {'loss': 0.7898, 'grad_norm': 0.7283092322406449, 'learning_rate': 2e-05, 'epoch': 0.84}
+0: {'loss': 0.7732, 'grad_norm': 0.6949231445766547, 'learning_rate': 2e-05, 'epoch': 0.84}
+0: .82s/it]
                                                         

 84%|████████▎ | 12310/14725 [6:15:34<1:13:27,  1.82s/it]
 84%|████████▎ | 12311/14725 [6:15:36<1:12:34,  1.80s/it]
 84%|████████▎ | 12312/14725 [6:15:37<1:12:14,  1.80s/it]
 84%|████████▎ | 12313/14725 [6:15:39<1:12:07,  1.79s/it]
 84%|████████▎ | 12314/14725 [6:15:41<1:11:36,  1.78s/it]
 84%|████████▎ | 12315/14725 [6:15:43<1:12:24,  1.80s/it]
 84%|████████▎ | 12316/14725 [6:15:45<1:12:29,  1.81s/it]
 84%|████████▎ | 12317/14725 [6:15:46<1:12:09,  1.80s/it]
 84%|████████▎ | 12318/14725 [6:15:48<1:11:51,  1.79s/it]
 84%|████████▎ | 12319/14725 [6:15:50<1:12:16,  1.80s/it]
 84%|████████▎ | 12320/14725 [6:15:52<1:13:47,  1.84s/it]
                                                         

 84%|████████▎ | 12320/14725 [6:15:52<1:13:
+0: {'loss': 0.8017, 'grad_norm': 0.6741224195965433, 'learning_rate': 2e-05, 'epoch': 0.84}
+0: 47,  1.84s/it]
 84%|████████▎ | 12321/14725 [6:15:54<1:12:43,  1.81s/it]
 84%|████████▎ | 12322/14725 [6:15:55<1:12:55,  1.82s/it]
 84%|████████▎ | 12323/14725 [6:15:57<1:12:15,  1.81s/it]
 84%|████████▎ | 12324/14725 [6:15:59<1:12:27,  1.81s/it]
 84%|████████▎ | 12325/14725 [6:16:01<1:12:39,  1.82s/it]
 84%|████████▎ | 12326/14725 [6:16:03<1:11:49,  1.80s/it]
 84%|████████▎ | 12327/14725 [6:16:04<1:11:17,  1.78s/it]
 84%|████████▎ | 12328/14725 [6:16:06<1:11:02,  1.78s/it]
 84%|████████▎ | 12329/14725 [6:16:08<1:11:11,  1.78s/it]
 84%|████████▎ | 12330/14725 [6:16:10<1:11:08,  1.78s/it]
                                                         

 84%|████████▎ | 12330/14725 [6:16:10<1:11:08,  1.78s/it]
 84%|████████▎ | 12331/14725 [6:16:12<1:11:56,  1.80s/it]
 84%|████████▎ | 123
+0: {'loss': 0.7892, 'grad_norm': 0.6516114716880111, 'learning_rate': 2e-05, 'epoch': 0.84}
+0: 32/14725 [6:16:13<1:11:41,  1.80s/it]
 84%|████████▍ | 12333/14725 [6:16:15<1:10:58,  1.78s/it]
 84%|████████▍ | 12334/14725 [6:16:17<1:12:01,  1.81s/it]
 84%|████████▍ | 12335/14725 [6:16:19<1:12:40,  1.82s/it]
 84%|████████▍ | 12336/14725 [6:16:21<1:11:55,  1.81s/it]
 84%|████████▍ | 12337/14725 [6:16:22<1:11:51,  1.81s/it]
 84%|████████▍ | 12338/14725 [6:16:24<1:11:49,  1.81s/it]
 84%|████████▍ | 12339/14725 [6:16:26<1:11:40,  1.80s/it]
 84%|████████▍ | 12340/14725 [6:16:28<1:11:50,  1.81s/it]
                                                         

 84%|████████▍ | 12340/14725 [6:16:28<1:11:50,  1.81s/it]
 84%|████████▍ | 12341/14725 [6:16:30<1:11:33,  1.80s/it]
 84%|████████▍ | 12342/14725 [6:16:31<1:11:51,  1.81s/it]
 84%|████████▍ | 12343/14725 [6:16:33<1:12:17,  1.82s/it]
 84%|███�
+0: {'loss': 0.8059, 'grad_norm': 0.6616537140138739, 'learning_rate': 2e-05, 'epoch': 0.84}
+0: ��████▍ | 12344/14725 [6:16:35<1:11:40,  1.81s/it]
 84%|████████▍ | 12345/14725 [6:16:37<1:11:17,  1.80s/it]
 84%|████████▍ | 12346/14725 [6:16:39<1:11:03,  1.79s/it]
 84%|████████▍ | 12347/14725 [6:16:40<1:11:06,  1.79s/it]
 84%|████████▍ | 12348/14725 [6:16:42<1:11:31,  1.81s/it]
 84%|████████▍ | 12349/14725 [6:16:44<1:11:35,  1.81s/it]
 84%|████████▍ | 12350/14725 [6:16:46<1:11:55,  1.82s/it]
                                                         

 84%|████████▍ | 12350/14725 [6:16:46<1:11:55,  1.82s/it]
 84%|████████▍ | 12351/14725 [6:16:48<1:12:28,  1.83s/it]
 84%|████████▍ | 12352/14725 [6:16:50<1:12:34,  1.84s/it]
 84%|████████▍ | 12353/14725 [6:16:51<1:11:55,  1.82s/it]
 84%|████████▍ | 12354/14725 [6:16:53<1:12:21,  1.83s/it]
 84%|████████▍ | 12355/14725 [6:16:55<1:12:31,  1.
+0: {'loss': 0.7908, 'grad_norm': 0.7363392444751493, 'learning_rate': 2e-05, 'epoch': 0.84}
+0: 84s/it]
 84%|████████▍ | 12356/14725 [6:16:57<1:12:14,  1.83s/it]
 84%|████████▍ | 12357/14725 [6:16:59<1:12:16,  1.83s/it]
 84%|████████▍ | 12358/14725 [6:17:01<1:13:28,  1.86s/it]
 84%|████████▍ | 12359/14725 [6:17:03<1:12:54,  1.85s/it]
 84%|████████▍ | 12360/14725 [6:17:04<1:12:35,  1.84s/it]
                                                         

 84%|████████▍ | 12360/14725 [6:17:04<1:12:35,  1.84s/it]
 84%|████████▍ | 12361/14725 [6:17:06<1:12:13,  1.83s/it]
 84%|████████▍ | 12362/14725 [6:17:08<1:11:33,  1.82s/it]
 84%|████████▍ | 12363/14725 [6:17:10<1:11:15,  1.81s/it]
 84%|████████▍ | 12364/14725 [6:17:12<1:11:51,  1.83s/it]
 84%|████████▍ | 12365/14725 [6:17:13<1:11:57,  1.83s/it]
 84%|████████▍ | 12366/14725 [6:17:15<1:11:18,  1.81s/it]
 84%|████████▍ | 12367/1472
+0: {'loss': 0.7966, 'grad_norm': 0.6522726543287205, 'learning_rate': 2e-05, 'epoch': 0.84}
+0: 5 [6:17:17<1:11:29,  1.82s/it]
 84%|████████▍ | 12368/14725 [6:17:19<1:11:08,  1.81s/it]
 84%|████████▍ | 12369/14725 [6:17:21<1:11:01,  1.81s/it]
 84%|████████▍ | 12370/14725 [6:17:22<1:11:01,  1.81s/it]
                                                         

 84%|████████▍ | 12370/14725 [6:17:22<1:11:01,  1.81s/it]
 84%|████████▍ | 12371/14725 [6:17:24<1:11:16,  1.82s/it]
 84%|████████▍ | 12372/14725 [6:17:26<1:11:04,  1.81s/it]
 84%|████████▍ | 12373/14725 [6:17:28<1:11:13,  1.82s/it]
 84%|████████▍ | 12374/14725 [6:17:30<1:11:02,  1.81s/it]
 84%|████████▍ | 12375/14725 [6:17:31<1:10:27,  1.80s/it]
 84%|████████▍ | 12376/14725 [6:17:33<1:10:17,  1.80s/it]
 84%|████████▍ | 12377/14725 [6:17:35<1:10:47,  1.81s/it]
 84%|████████▍ | 12378/14725 [6:17:37<1:11:00,  1.82s/it]
 84%|█████�
+0: {'loss': 0.782, 'grad_norm': 0.7121454389977415, 'learning_rate': 2e-05, 'epoch': 0.84}
+0: �██▍ | 12379/14725 [6:17:39<1:10:31,  1.80s/it]
 84%|████████▍ | 12380/14725 [6:17:41<1:10:55,  1.81s/it]
                                                         

 84%|████████▍ | 12380/14725 [6:17:41<1:10:55,  1.81s/it]
 84%|████████▍ | 12381/14725 [6:17:42<1:11:13,  1.82s/it]
 84%|████████▍ | 12382/14725 [6:17:44<1:11:06,  1.82s/it]
 84%|████████▍ | 12383/14725 [6:17:46<1:10:59,  1.82s/it]
 84%|████████▍ | 12384/14725 [6:17:48<1:11:07,  1.82s/it]
 84%|████████▍ | 12385/14725 [6:17:50<1:10:25,  1.81s/it]
 84%|████████▍ | 12386/14725 [6:17:51<1:10:32,  1.81s/it]
 84%|████████▍ | 12387/14725 [6:17:53<1:10:50,  1.82s/it]
 84%|████████▍ | 12388/14725 [6:17:55<1:10:37,  1.81s/it]
 84%|████████▍ | 12389/14725 [6:17:57<1:11:38,  1.84s/it]
 84%|████████▍ | 12390/14725 [6:17:59<1:10:50,  1.82s/it]
+0: {'loss': 0.776, 'grad_norm': 0.6488206753720336, 'learning_rate': 2e-05, 'epoch': 0.84}
+0: {'loss': 0.7913, 'grad_norm': 0.6819136751344052, 'learning_rate': 2e-05, 'epoch': 0.84}
+0: 
                                                         

 84%|████████▍ | 12390/14725 [6:17:59<1:10:50,  1.82s/it]
 84%|████████▍ | 12391/14725 [6:18:01<1:10:43,  1.82s/it]
 84%|████████▍ | 12392/14725 [6:18:02<1:10:30,  1.81s/it]
 84%|████████▍ | 12393/14725 [6:18:04<1:10:38,  1.82s/it]
 84%|████████▍ | 12394/14725 [6:18:06<1:11:07,  1.83s/it]
 84%|████████▍ | 12395/14725 [6:18:08<1:12:06,  1.86s/it]
 84%|████████▍ | 12396/14725 [6:18:10<1:11:35,  1.84s/it]
 84%|████████▍ | 12397/14725 [6:18:12<1:11:43,  1.85s/it]
 84%|████████▍ | 12398/14725 [6:18:13<1:11:24,  1.84s/it]
 84%|████████▍ | 12399/14725 [6:18:15<1:10:44,  1.82s/it]
 84%|████████▍ | 12400/14725 [6:18:17<1:10:04,  1.81s/it]
                                                         

 84%|████████▍ | 12400/14725 [6:18:17<1:10:04,  1.8
+0: {'loss': 0.7967, 'grad_norm': 0.6756220759824076, 'learning_rate': 2e-05, 'epoch': 0.84}
+0: 1s/it]
 84%|████████▍ | 12401/14725 [6:18:19<1:09:31,  1.80s/it]
 84%|████████▍ | 12402/14725 [6:18:21<1:10:47,  1.83s/it]
 84%|████████▍ | 12403/14725 [6:18:23<1:10:54,  1.83s/it]
 84%|████████▍ | 12404/14725 [6:18:24<1:10:10,  1.81s/it]
 84%|████████▍ | 12405/14725 [6:18:26<1:09:29,  1.80s/it]
 84%|████████▍ | 12406/14725 [6:18:28<1:09:12,  1.79s/it]
 84%|████████▍ | 12407/14725 [6:18:30<1:09:33,  1.80s/it]
 84%|████████▍ | 12408/14725 [6:18:31<1:09:11,  1.79s/it]
 84%|████████▍ | 12409/14725 [6:18:33<1:09:23,  1.80s/it]
 84%|████████▍ | 12410/14725 [6:18:35<1:08:33,  1.78s/it]
                                                         

 84%|████████▍ | 12410/14725 [6:18:35<1:08:33,  1.78s/it]
 84%|████████▍ | 12411/14725 [6:18:37<1:08:28,  1.78s/it]
 84%|████████▍ | 12412/14725
+0: {'loss': 0.8168, 'grad_norm': 0.6709738496734552, 'learning_rate': 2e-05, 'epoch': 0.84}
+0:  [6:18:39<1:08:56,  1.79s/it]
 84%|████████▍ | 12413/14725 [6:18:40<1:09:10,  1.80s/it]
 84%|████████▍ | 12414/14725 [6:18:42<1:09:12,  1.80s/it]
 84%|████████▍ | 12415/14725 [6:18:44<1:09:43,  1.81s/it]
 84%|████████▍ | 12416/14725 [6:18:46<1:09:04,  1.79s/it]
 84%|████████▍ | 12417/14725 [6:18:48<1:08:37,  1.78s/it]
 84%|████████▍ | 12418/14725 [6:18:49<1:08:10,  1.77s/it]
 84%|████████▍ | 12419/14725 [6:18:51<1:08:09,  1.77s/it]
 84%|████████▍ | 12420/14725 [6:18:53<1:08:03,  1.77s/it]
                                                         

 84%|████████▍ | 12420/14725 [6:18:53<1:08:03,  1.77s/it]
 84%|████████▍ | 12421/14725 [6:18:55<1:07:56,  1.77s/it]
 84%|████████▍ | 12422/14725 [6:18:56<1:07:59,  1.77s/it]
 84%|████████▍ | 12423/14725 [6:18:58<1:08:18,  1.78s/it]
 84%|██████
+0: {'loss': 0.8038, 'grad_norm': 0.6615343304813699, 'learning_rate': 2e-05, 'epoch': 0.84}
+0: ██▍ | 12424/14725 [6:19:00<1:08:46,  1.79s/it]
 84%|████████▍ | 12425/14725 [6:19:02<1:08:34,  1.79s/it]
 84%|████████▍ | 12426/14725 [6:19:04<1:08:43,  1.79s/it]
 84%|████████▍ | 12427/14725 [6:19:05<1:09:51,  1.82s/it]
 84%|████████▍ | 12428/14725 [6:19:07<1:09:29,  1.82s/it]
 84%|████████▍ | 12429/14725 [6:19:09<1:08:48,  1.80s/it]
 84%|████████▍ | 12430/14725 [6:19:11<1:09:28,  1.82s/it]
                                                         

 84%|████████▍ | 12430/14725 [6:19:11<1:09:28,  1.82s/it]
 84%|████████▍ | 12431/14725 [6:19:13<1:09:58,  1.83s/it]
 84%|████████▍ | 12432/14725 [6:19:15<1:10:04,  1.83s/it]
 84%|████████▍ | 12433/14725 [6:19:16<1:09:21,  1.82s/it]
 84%|████████▍ | 12434/14725 [6:19:18<1:09:44,  1.83s/it]
 84%|████████▍ | 12435/14725 [6:19:20<1:09:39,  1.83s/it]
+0: {'loss': 0.7789, 'grad_norm': 0.7027214985060048, 'learning_rate': 2e-05, 'epoch': 0.84}
+0:  84%|████████▍ | 12436/14725 [6:19:22<1:08:33,  1.80s/it]
 84%|████████▍ | 12437/14725 [6:19:24<1:08:30,  1.80s/it]
 84%|████████▍ | 12438/14725 [6:19:25<1:08:24,  1.79s/it]
 84%|████████▍ | 12439/14725 [6:19:27<1:08:30,  1.80s/it]
 84%|████████▍ | 12440/14725 [6:19:29<1:09:17,  1.82s/it]
                                                         

 84%|████████▍ | 12440/14725 [6:19:29<1:09:17,  1.82s/it]
 84%|████████▍ | 12441/14725 [6:19:31<1:08:45,  1.81s/it]
 84%|████████▍ | 12442/14725 [6:19:33<1:09:38,  1.83s/it]
 85%|████████▍ | 12443/14725 [6:19:34<1:09:16,  1.82s/it]
 85%|████████▍ | 12444/14725 [6:19:36<1:09:06,  1.82s/it]
 85%|████████▍ | 12445/14725 [6:19:38<1:10:06,  1.84s/it]
 85%|████████▍ | 12446/14725 [6:19:40<1:08:56,  1.82s/it]
 85%|████████▍ | 12447/14725 [6:19:
+0: {'loss': 0.8027, 'grad_norm': 0.7241670852068708, 'learning_rate': 2e-05, 'epoch': 0.85}
+0: 42<1:08:19,  1.80s/it]
 85%|████████▍ | 12448/14725 [6:19:44<1:08:18,  1.80s/it]
 85%|████████▍ | 12449/14725 [6:19:45<1:08:33,  1.81s/it]
 85%|████████▍ | 12450/14725 [6:19:47<1:08:33,  1.81s/it]
                                                         

 85%|████████▍ | 12450/14725 [6:19:47<1:08:33,  1.81s/it]
 85%|████████▍ | 12451/14725 [6:19:49<1:07:59,  1.79s/it]
 85%|████████▍ | 12452/14725 [6:19:51<1:07:17,  1.78s/it]
 85%|████████▍ | 12453/14725 [6:19:52<1:07:08,  1.77s/it]
 85%|████████▍ | 12454/14725 [6:19:54<1:06:51,  1.77s/it]
 85%|████████▍ | 12455/14725 [6:19:56<1:06:38,  1.76s/it]
 85%|████████▍ | 12456/14725 [6:19:58<1:08:44,  1.82s/it]
 85%|████████▍ | 12457/14725 [6:20:00<1:08:23,  1.81s/it]
 85%|████████▍ | 12458/14725 [6:20:01<1:08:25,  1.81s/it]
 85%|████████�
+0: {'loss': 0.7947, 'grad_norm': 0.6601879016007849, 'learning_rate': 2e-05, 'epoch': 0.85}
+0: �� | 12459/14725 [6:20:03<1:09:11,  1.83s/it]
 85%|████████▍ | 12460/14725 [6:20:05<1:08:31,  1.82s/it]
                                                         

 85%|████████▍ | 12460/14725 [6:20:05<1:08:31,  1.82s/it]
 85%|████████▍ | 12461/14725 [6:20:07<1:08:35,  1.82s/it]
 85%|████████▍ | 12462/14725 [6:20:09<1:08:49,  1.82s/it]
 85%|████████▍ | 12463/14725 [6:20:11<1:08:03,  1.81s/it]
 85%|████████▍ | 12464/14725 [6:20:12<1:08:34,  1.82s/it]
 85%|████████▍ | 12465/14725 [6:20:14<1:08:32,  1.82s/it]
 85%|████████▍ | 12466/14725 [6:20:16<1:07:59,  1.81s/it]
 85%|████████▍ | 12467/14725 [6:20:18<1:07:56,  1.81s/it]
 85%|████████▍ | 12468/14725 [6:20:20<1:08:03,  1.81s/it]
 85%|████████▍ | 12469/14725 [6:20:21<1:08:47,  1.83s/it]
 85%|████████▍ | 12470/14725 [6:20:23<1:07:50,  1.80s/it]
       
+0: {'loss': 0.7964, 'grad_norm': 0.6775246620782922, 'learning_rate': 2e-05, 'epoch': 0.85}
+0: {'loss': 0.7973, 'grad_norm': 0.6723740962266143, 'learning_rate': 2e-05, 'epoch': 0.85}
+0:                                                   

 85%|████████▍ | 12470/14725 [6:20:23<1:07:50,  1.80s/it]
 85%|████████▍ | 12471/14725 [6:20:25<1:08:05,  1.81s/it]
 85%|████████▍ | 12472/14725 [6:20:27<1:08:25,  1.82s/it]
 85%|████████▍ | 12473/14725 [6:20:29<1:08:46,  1.83s/it]
 85%|████████▍ | 12474/14725 [6:20:31<1:09:02,  1.84s/it]
 85%|████████▍ | 12475/14725 [6:20:32<1:08:46,  1.83s/it]
 85%|████████▍ | 12476/14725 [6:20:34<1:09:32,  1.86s/it]
 85%|████████▍ | 12477/14725 [6:20:36<1:09:38,  1.86s/it]
 85%|████████▍ | 12478/14725 [6:20:38<1:08:48,  1.84s/it]
 85%|████████▍ | 12479/14725 [6:20:40<1:08:55,  1.84s/it]
 85%|████████▍ | 12480/14725 [6:20:42<1:08:47,  1.84s/it]
                                                         

 85%|████████▍ | 12480/14725 [6:20:42<1:08:47,  1.84s/it]
 
+0: {'loss': 0.8104, 'grad_norm': 0.6377315160701962, 'learning_rate': 2e-05, 'epoch': 0.85}
+0: 85%|████████▍ | 12481/14725 [6:20:44<1:08:30,  1.83s/it]
 85%|████████▍ | 12482/14725 [6:20:45<1:07:48,  1.81s/it]
 85%|████████▍ | 12483/14725 [6:20:47<1:07:49,  1.82s/it]
 85%|████████▍ | 12484/14725 [6:20:49<1:07:23,  1.80s/it]
 85%|████████▍ | 12485/14725 [6:20:51<1:07:29,  1.81s/it]
 85%|████████▍ | 12486/14725 [6:20:52<1:07:01,  1.80s/it]
 85%|████████▍ | 12487/14725 [6:20:54<1:07:29,  1.81s/it]
 85%|████████▍ | 12488/14725 [6:20:56<1:07:41,  1.82s/it]
 85%|████████▍ | 12489/14725 [6:20:58<1:07:40,  1.82s/it]
 85%|████████▍ | 12490/14725 [6:21:00<1:08:37,  1.84s/it]
                                                         

 85%|████████▍ | 12490/14725 [6:21:00<1:08:37,  1.84s/it]
 85%|████████▍ | 12491/14725 [6:21:02<1:08:15,  1.83s/it]
 85%|████████▍ | 12492/14725 [6:21:0
+0: {'loss': 0.797, 'grad_norm': 0.6718838145266478, 'learning_rate': 2e-05, 'epoch': 0.85}
+0: 3<1:07:19,  1.81s/it]
 85%|████████▍ | 12493/14725 [6:21:05<1:07:12,  1.81s/it]
 85%|████████▍ | 12494/14725 [6:21:07<1:06:48,  1.80s/it]
 85%|████████▍ | 12495/14725 [6:21:09<1:06:27,  1.79s/it]
 85%|████████▍ | 12496/14725 [6:21:11<1:06:35,  1.79s/it]
 85%|████████▍ | 12497/14725 [6:21:12<1:06:00,  1.78s/it]
 85%|████████▍ | 12498/14725 [6:21:14<1:07:07,  1.81s/it]
 85%|████████▍ | 12499/14725 [6:21:16<1:07:39,  1.82s/it]
 85%|████████▍ | 12500/14725 [6:21:18<1:09:02,  1.86s/it]
                                                         

 85%|████████▍ | 12500/14725 [6:21:18<1:09:02,  1.86s/it]
 85%|████████▍ | 12501/14725 [6:21:20<1:08:44,  1.85s/it]
 85%|████████▍ | 12502/14725 [6:21:22<1:08:23,  1.85s/it]
 85%|████████▍ | 12503/14725 [6:21:23<1:07:45,  1.83s/it]
 85%|████████�
+0: {'loss': 0.8053, 'grad_norm': 0.6655184358385309, 'learning_rate': 2e-05, 'epoch': 0.85}
+0: � | 12504/14725 [6:21:25<1:07:53,  1.83s/it]
 85%|████████▍ | 12505/14725 [6:21:27<1:06:54,  1.81s/it]
 85%|████████▍ | 12506/14725 [6:21:29<1:06:57,  1.81s/it]
 85%|████████▍ | 12507/14725 [6:21:31<1:08:08,  1.84s/it]
 85%|████████▍ | 12508/14725 [6:21:33<1:08:07,  1.84s/it]
 85%|████████▍ | 12509/14725 [6:21:34<1:07:44,  1.83s/it]
 85%|████████▍ | 12510/14725 [6:21:36<1:08:27,  1.85s/it]
                                                         

 85%|████████▍ | 12510/14725 [6:21:36<1:08:27,  1.85s/it]
 85%|████████▍ | 12511/14725 [6:21:38<1:07:58,  1.84s/it]
 85%|████████▍ | 12512/14725 [6:21:40<1:07:08,  1.82s/it]
 85%|████████▍ | 12513/14725 [6:21:42<1:06:58,  1.82s/it]
 85%|████████▍ | 12514/14725 [6:21:44<1:06:42,  1.81s/it]
 85%|████████▍ | 12515/14725 [6:21:45<1:06:46,  1.81s/it]
 85%|█
+0: {'loss': 0.7777, 'grad_norm': 0.6714617440660752, 'learning_rate': 2e-05, 'epoch': 0.85}
+0: ███████▍ | 12516/14725 [6:21:47<1:06:11,  1.80s/it]
 85%|████████▌ | 12517/14725 [6:21:49<1:05:59,  1.79s/it]
 85%|████████▌ | 12518/14725 [6:21:51<1:06:19,  1.80s/it]
 85%|████████▌ | 12519/14725 [6:21:52<1:05:44,  1.79s/it]
 85%|████████▌ | 12520/14725 [6:21:54<1:05:57,  1.79s/it]
                                                         

 85%|████████▌ | 12520/14725 [6:21:54<1:05:57,  1.79s/it]
 85%|████████▌ | 12521/14725 [6:21:56<1:06:20,  1.81s/it]
 85%|████████▌ | 12522/14725 [6:21:58<1:06:48,  1.82s/it]
 85%|████████▌ | 12523/14725 [6:22:00<1:06:27,  1.81s/it]
 85%|████████▌ | 12524/14725 [6:22:02<1:06:52,  1.82s/it]
 85%|████████▌ | 12525/14725 [6:22:03<1:07:19,  1.84s/it]
 85%|████████▌ | 12526/14725 [6:22:05<1:06:13,  1.81s/it]
 85%|████████▌ | 12527/14725 [6:22:07<1:06:
+0: {'loss': 0.803, 'grad_norm': 0.6789865384497815, 'learning_rate': 2e-05, 'epoch': 0.85}
+0: 47,  1.82s/it]
 85%|████████▌ | 12528/14725 [6:22:09<1:06:13,  1.81s/it]
 85%|████████▌ | 12529/14725 [6:22:11<1:05:45,  1.80s/it]
 85%|████████▌ | 12530/14725 [6:22:12<1:05:21,  1.79s/it]
                                                         

 85%|████████▌ | 12530/14725 [6:22:12<1:05:21,  1.79s/it]
 85%|████████▌ | 12531/14725 [6:22:14<1:06:05,  1.81s/it]
 85%|████████▌ | 12532/14725 [6:22:16<1:06:00,  1.81s/it]
 85%|████████▌ | 12533/14725 [6:22:18<1:06:07,  1.81s/it]
 85%|████████▌ | 12534/14725 [6:22:20<1:06:31,  1.82s/it]
 85%|████████▌ | 12535/14725 [6:22:22<1:06:18,  1.82s/it]
 85%|████████▌ | 12536/14725 [6:22:23<1:05:41,  1.80s/it]
 85%|████████▌ | 12537/14725 [6:22:25<1:06:45,  1.83s/it]
 85%|████████▌ | 12538/14725 [6:22:27<1:06:27,  1.82s/it]
 85%|████████▌ | 125
+0: {'loss': 0.7771, 'grad_norm': 0.6435270333210501, 'learning_rate': 2e-05, 'epoch': 0.85}
+0: 39/14725 [6:22:29<1:06:40,  1.83s/it]
 85%|████████▌ | 12540/14725 [6:22:31<1:05:53,  1.81s/it]
                                                         

 85%|████████▌ | 12540/14725 [6:22:31<1:05:53,  1.81s/it]
 85%|████████▌ | 12541/14725 [6:22:32<1:06:07,  1.82s/it]
 85%|████████▌ | 12542/14725 [6:22:34<1:06:11,  1.82s/it]
 85%|████████▌ | 12543/14725 [6:22:36<1:06:13,  1.82s/it]
 85%|████████▌ | 12544/14725 [6:22:38<1:05:38,  1.81s/it]
 85%|████████▌ | 12545/14725 [6:22:40<1:05:50,  1.81s/it]
 85%|████████▌ | 12546/14725 [6:22:41<1:05:32,  1.80s/it]
 85%|████████▌ | 12547/14725 [6:22:43<1:05:21,  1.80s/it]
 85%|████████▌ | 12548/14725 [6:22:45<1:05:14,  1.80s/it]
 85%|████████▌ | 12549/14725 [6:22:47<1:04:46,  1.79s/it]
 85%|████████▌ | 12550/14725 [6:22:49<1:04:44,  1.79s/it]
               
+0: {'loss': 0.7948, 'grad_norm': 0.6512451015848852, 'learning_rate': 2e-05, 'epoch': 0.85}
+0: {'loss': 0.7747, 'grad_norm': 0.709508992101671, 'learning_rate': 2e-05, 'epoch': 0.85}
+0:                                           

 85%|████████▌ | 12550/14725 [6:22:49<1:04:44,  1.79s/it]
 85%|████████▌ | 12551/14725 [6:22:50<1:05:44,  1.81s/it]
 85%|████████▌ | 12552/14725 [6:22:52<1:05:01,  1.80s/it]
 85%|████████▌ | 12553/14725 [6:22:54<1:05:12,  1.80s/it]
 85%|████████▌ | 12554/14725 [6:22:56<1:05:28,  1.81s/it]
 85%|████████▌ | 12555/14725 [6:22:58<1:05:11,  1.80s/it]
 85%|████████▌ | 12556/14725 [6:22:59<1:05:17,  1.81s/it]
 85%|████████▌ | 12557/14725 [6:23:01<1:04:52,  1.80s/it]
 85%|████████▌ | 12558/14725 [6:23:03<1:05:56,  1.83s/it]
 85%|████████▌ | 12559/14725 [6:23:05<1:06:33,  1.84s/it]
 85%|████████▌ | 12560/14725 [6:23:07<1:06:14,  1.84s/it]
                                                         

 85%|████████▌ | 12560/14725 [6:23:07<1:06:14,  1.84s/it]
 85%|█�
+0: {'loss': 0.806, 'grad_norm': 0.6690648611229569, 'learning_rate': 2e-05, 'epoch': 0.85}
+0: ��██████▌ | 12561/14725 [6:23:09<1:05:38,  1.82s/it]
 85%|████████▌ | 12562/14725 [6:23:10<1:04:50,  1.80s/it]
 85%|████████▌ | 12563/14725 [6:23:12<1:05:00,  1.80s/it]
 85%|████████▌ | 12564/14725 [6:23:14<1:04:17,  1.79s/it]
 85%|████████▌ | 12565/14725 [6:23:16<1:04:25,  1.79s/it]
 85%|████████▌ | 12566/14725 [6:23:17<1:04:04,  1.78s/it]
 85%|████████▌ | 12567/14725 [6:23:19<1:04:10,  1.78s/it]
 85%|████████▌ | 12568/14725 [6:23:21<1:04:46,  1.80s/it]
 85%|████████▌ | 12569/14725 [6:23:23<1:04:45,  1.80s/it]
 85%|████████▌ | 12570/14725 [6:23:25<1:04:09,  1.79s/it]
                                                         

 85%|████████▌ | 12570/14725 [6:23:25<1:04:09,  1.79s/it]
 85%|████████▌ | 12571/14725 [6:23:26<1:04:13,  1.79s/it]
 85%|████████▌ | 12572/14725 [6:23:28<1:04:1
+0: {'loss': 0.7997, 'grad_norm': 0.6482930609147153, 'learning_rate': 2e-05, 'epoch': 0.85}
+0: 6,  1.79s/it]
 85%|████████▌ | 12573/14725 [6:23:30<1:04:31,  1.80s/it]
 85%|████████▌ | 12574/14725 [6:23:32<1:06:19,  1.85s/it]
 85%|████████▌ | 12575/14725 [6:23:34<1:06:54,  1.87s/it]
 85%|████████▌ | 12576/14725 [6:23:36<1:05:52,  1.84s/it]
 85%|████████▌ | 12577/14725 [6:23:38<1:05:10,  1.82s/it]
 85%|████████▌ | 12578/14725 [6:23:39<1:06:21,  1.85s/it]
 85%|████████▌ | 12579/14725 [6:23:41<1:05:30,  1.83s/it]
 85%|████████▌ | 12580/14725 [6:23:43<1:05:16,  1.83s/it]
                                                         

 85%|████████▌ | 12580/14725 [6:23:43<1:05:16,  1.83s/it]
 85%|████████▌ | 12581/14725 [6:23:45<1:04:38,  1.81s/it]
 85%|████████▌ | 12582/14725 [6:23:47<1:04:07,  1.80s/it]
 85%|████████▌ | 12583/14725 [6:23:48<1:03:33,  1.78s/it]
 85%|████████▌ | 1258
+0: {'loss': 0.7791, 'grad_norm': 0.7197360088383964, 'learning_rate': 2e-05, 'epoch': 0.85}
+0: 4/14725 [6:23:50<1:03:21,  1.78s/it]
 85%|████████▌ | 12585/14725 [6:23:52<1:03:40,  1.79s/it]
 85%|████████▌ | 12586/14725 [6:23:54<1:05:01,  1.82s/it]
 85%|████████▌ | 12587/14725 [6:23:56<1:06:06,  1.86s/it]
 85%|████████▌ | 12588/14725 [6:23:57<1:05:00,  1.83s/it]
 85%|████████▌ | 12589/14725 [6:23:59<1:04:15,  1.81s/it]
 86%|████████▌ | 12590/14725 [6:24:01<1:04:34,  1.81s/it]
                                                         

 86%|████████▌ | 12590/14725 [6:24:01<1:04:34,  1.81s/it]
 86%|████████▌ | 12591/14725 [6:24:03<1:03:48,  1.79s/it]
 86%|████████▌ | 12592/14725 [6:24:05<1:03:41,  1.79s/it]
 86%|████████▌ | 12593/14725 [6:24:06<1:03:50,  1.80s/it]
 86%|████████▌ | 12594/14725 [6:24:08<1:03:55,  1.80s/it]
 86%|████████▌ | 12595/14725 [6:24:10<1:04:08,  1.81s/it]
 86%|███�
+0: {'loss': 0.8014, 'grad_norm': 0.6516802364731497, 'learning_rate': 2e-05, 'epoch': 0.86}
+0: �████▌ | 12596/14725 [6:24:12<1:04:11,  1.81s/it]
 86%|████████▌ | 12597/14725 [6:24:14<1:04:24,  1.82s/it]
 86%|████████▌ | 12598/14725 [6:24:16<1:04:34,  1.82s/it]
 86%|████████▌ | 12599/14725 [6:24:17<1:04:46,  1.83s/it]
 86%|████████▌ | 12600/14725 [6:24:19<1:04:08,  1.81s/it]
                                                         

 86%|████████▌ | 12600/14725 [6:24:19<1:04:08,  1.81s/it]
 86%|████████▌ | 12601/14725 [6:24:21<1:05:18,  1.84s/it]
 86%|████████▌ | 12602/14725 [6:24:23<1:05:07,  1.84s/it]
 86%|████████▌ | 12603/14725 [6:24:25<1:05:02,  1.84s/it]
 86%|████████▌ | 12604/14725 [6:24:27<1:04:47,  1.83s/it]
 86%|████████▌ | 12605/14725 [6:24:28<1:04:30,  1.83s/it]
 86%|████████▌ | 12606/14725 [6:24:30<1:04:14,  1.82s/it]
 86%|████████▌ | 12607/14725 [6:24:32<1:03:22,  1.8
+0: {'loss': 0.8044, 'grad_norm': 0.7196742691650431, 'learning_rate': 2e-05, 'epoch': 0.86}
+0: 0s/it]
 86%|████████▌ | 12608/14725 [6:24:34<1:03:17,  1.79s/it]
 86%|████████▌ | 12609/14725 [6:24:36<1:03:34,  1.80s/it]
 86%|████████▌ | 12610/14725 [6:24:37<1:03:15,  1.79s/it]
                                                         

 86%|████████▌ | 12610/14725 [6:24:37<1:03:15,  1.79s/it]
 86%|████████▌ | 12611/14725 [6:24:39<1:03:17,  1.80s/it]
 86%|████████▌ | 12612/14725 [6:24:41<1:03:27,  1.80s/it]
 86%|████████▌ | 12613/14725 [6:24:43<1:04:33,  1.83s/it]
 86%|████████▌ | 12614/14725 [6:24:45<1:04:19,  1.83s/it]
 86%|████████▌ | 12615/14725 [6:24:46<1:03:58,  1.82s/it]
 86%|████████▌ | 12616/14725 [6:24:48<1:03:38,  1.81s/it]
 86%|████████▌ | 12617/14725 [6:24:50<1:03:19,  1.80s/it]
 86%|████████▌ | 12618/14725 [6:24:52<1:02:56,  1.79s/it]
 86%|████████▌ | 12619/14725
+0: {'loss': 0.7851, 'grad_norm': 0.6781399998458738, 'learning_rate': 2e-05, 'epoch': 0.86}
+0:  [6:24:54<1:02:36,  1.78s/it]
 86%|████████▌ | 12620/14725 [6:24:55<1:02:10,  1.77s/it]
                                                         

 86%|████████▌ | 12620/14725 [6:24:55<1:02:10,  1.77s/it]
 86%|████████▌ | 12621/14725 [6:24:57<1:02:15,  1.78s/it]
 86%|████████▌ | 12622/14725 [6:24:59<1:02:42,  1.79s/it]
 86%|████████▌ | 12623/14725 [6:25:01<1:03:09,  1.80s/it]
 86%|████████▌ | 12624/14725 [6:25:02<1:02:39,  1.79s/it]
 86%|████████▌ | 12625/14725 [6:25:04<1:02:59,  1.80s/it]
 86%|████████▌ | 12626/14725 [6:25:06<1:03:16,  1.81s/it]
 86%|████████▌ | 12627/14725 [6:25:08<1:02:57,  1.80s/it]
 86%|████████▌ | 12628/14725 [6:25:10<1:03:01,  1.80s/it]
 86%|████████▌ | 12629/14725 [6:25:12<1:04:08,  1.84s/it]
 86%|████████▌ | 12630/14725 [6:25:13<1:03:57,  1.83s/it]
                       
+0: {'loss': 0.7834, 'grad_norm': 0.6839559250702767, 'learning_rate': 2e-05, 'epoch': 0.86}
+0: {'loss': 0.7799, 'grad_norm': 0.6294576402572964, 'learning_rate': 2e-05, 'epoch': 0.86}
+0:                                   

 86%|████████▌ | 12630/14725 [6:25:13<1:03:57,  1.83s/it]
 86%|████████▌ | 12631/14725 [6:25:15<1:04:16,  1.84s/it]
 86%|████████▌ | 12632/14725 [6:25:17<1:03:52,  1.83s/it]
 86%|████████▌ | 12633/14725 [6:25:19<1:02:57,  1.81s/it]
 86%|████████▌ | 12634/14725 [6:25:21<1:02:37,  1.80s/it]
 86%|████████▌ | 12635/14725 [6:25:22<1:02:20,  1.79s/it]
 86%|████████▌ | 12636/14725 [6:25:24<1:01:44,  1.77s/it]
 86%|████████▌ | 12637/14725 [6:25:26<1:02:14,  1.79s/it]
 86%|████████▌ | 12638/14725 [6:25:28<1:02:15,  1.79s/it]
 86%|████████▌ | 12639/14725 [6:25:30<1:02:16,  1.79s/it]
 86%|████████▌ | 12640/14725 [6:25:31<1:02:39,  1.80s/it]
                                                         

 86%|████████▌ | 12640/14725 [6:25:31<1:02:39,  1.80s/it]
 86%|████
+0: {'loss': 0.7787, 'grad_norm': 0.6930799602745853, 'learning_rate': 2e-05, 'epoch': 0.86}
+0: ████▌ | 12641/14725 [6:25:33<1:02:23,  1.80s/it]
 86%|████████▌ | 12642/14725 [6:25:35<1:02:01,  1.79s/it]
 86%|████████▌ | 12643/14725 [6:25:37<1:01:38,  1.78s/it]
 86%|████████▌ | 12644/14725 [6:25:39<1:01:58,  1.79s/it]
 86%|████████▌ | 12645/14725 [6:25:40<1:02:53,  1.81s/it]
 86%|████████▌ | 12646/14725 [6:25:42<1:02:14,  1.80s/it]
 86%|████████▌ | 12647/14725 [6:25:44<1:02:31,  1.81s/it]
 86%|████████▌ | 12648/14725 [6:25:46<1:03:09,  1.82s/it]
 86%|████████▌ | 12649/14725 [6:25:48<1:02:24,  1.80s/it]
 86%|████████▌ | 12650/14725 [6:25:49<1:02:22,  1.80s/it]
                                                         

 86%|████████▌ | 12650/14725 [6:25:49<1:02:22,  1.80s/it]
 86%|████████▌ | 12651/14725 [6:25:51<1:02:38,  1.81s/it]
 86%|████████▌ | 12652/14725 [6:25:53<1:02:45,  1.82
+0: {'loss': 0.8138, 'grad_norm': 0.6536972525880853, 'learning_rate': 2e-05, 'epoch': 0.86}
+0: s/it]
 86%|████████▌ | 12653/14725 [6:25:55<1:02:18,  1.80s/it]
 86%|████████▌ | 12654/14725 [6:25:57<1:02:58,  1.82s/it]
 86%|████████▌ | 12655/14725 [6:25:59<1:03:27,  1.84s/it]
 86%|████████▌ | 12656/14725 [6:26:00<1:02:52,  1.82s/it]
 86%|████████▌ | 12657/14725 [6:26:02<1:02:49,  1.82s/it]
 86%|████████▌ | 12658/14725 [6:26:04<1:02:17,  1.81s/it]
 86%|████████▌ | 12659/14725 [6:26:06<1:02:15,  1.81s/it]
 86%|████████▌ | 12660/14725 [6:26:08<1:02:31,  1.82s/it]
                                                         

 86%|████████▌ | 12660/14725 [6:26:08<1:02:31,  1.82s/it]
 86%|████████▌ | 12661/14725 [6:26:09<1:02:08,  1.81s/it]
 86%|████████▌ | 12662/14725 [6:26:11<1:02:18,  1.81s/it]
 86%|████████▌ | 12663/14725 [6:26:13<1:02:29,  1.82s/it]
 86%|████████▌ | 12664/14725 
+0: {'loss': 0.7972, 'grad_norm': 0.6477735630161686, 'learning_rate': 2e-05, 'epoch': 0.86}
+0: [6:26:15<1:02:21,  1.82s/it]
 86%|████████▌ | 12665/14725 [6:26:17<1:01:45,  1.80s/it]
 86%|████████▌ | 12666/14725 [6:26:18<1:02:01,  1.81s/it]
 86%|████████▌ | 12667/14725 [6:26:20<1:02:56,  1.83s/it]
 86%|████████▌ | 12668/14725 [6:26:22<1:03:34,  1.85s/it]
 86%|████████▌ | 12669/14725 [6:26:24<1:02:49,  1.83s/it]
 86%|████████▌ | 12670/14725 [6:26:26<1:01:53,  1.81s/it]
                                                         

 86%|████████▌ | 12670/14725 [6:26:26<1:01:53,  1.81s/it]
 86%|████████▌ | 12671/14725 [6:26:28<1:01:38,  1.80s/it]
 86%|████████▌ | 12672/14725 [6:26:29<1:01:36,  1.80s/it]
 86%|████████▌ | 12673/14725 [6:26:31<1:01:19,  1.79s/it]
 86%|████████▌ | 12674/14725 [6:26:33<1:01:39,  1.80s/it]
 86%|████████▌ | 12675/14725 [6:26:35<1:01:18,  1.79s/it]
 86%|██████�
+0: {'loss': 0.7938, 'grad_norm': 0.6429567965799377, 'learning_rate': 2e-05, 'epoch': 0.86}
+0: ��█▌ | 12676/14725 [6:26:37<1:01:16,  1.79s/it]
 86%|████████▌ | 12677/14725 [6:26:38<1:00:55,  1.78s/it]
 86%|████████▌ | 12678/14725 [6:26:40<1:01:12,  1.79s/it]
 86%|████████▌ | 12679/14725 [6:26:42<1:01:18,  1.80s/it]
 86%|████████▌ | 12680/14725 [6:26:44<1:01:08,  1.79s/it]
                                                         

 86%|████████▌ | 12680/14725 [6:26:44<1:01:08,  1.79s/it]
 86%|████████▌ | 12681/14725 [6:26:46<1:01:31,  1.81s/it]
 86%|████████▌ | 12682/14725 [6:26:47<1:01:25,  1.80s/it]
 86%|██████���█▌ | 12683/14725 [6:26:49<1:01:38,  1.81s/it]
 86%|████████▌ | 12684/14725 [6:26:51<1:01:39,  1.81s/it]
 86%|████████▌ | 12685/14725 [6:26:53<1:01:37,  1.81s/it]
 86%|████████▌ | 12686/14725 [6:26:55<1:01:30,  1.81s/it]
 86%|████████▌ | 12687/14725 [6:26:56<1:01:25,  1.81s/it]
 
+0: {'loss': 0.7984, 'grad_norm': 0.6955224562879933, 'learning_rate': 2e-05, 'epoch': 0.86}
+0: 86%|████████▌ | 12688/14725 [6:26:58<1:01:58,  1.83s/it]
 86%|████████▌ | 12689/14725 [6:27:00<1:01:27,  1.81s/it]
 86%|████████▌ | 12690/14725 [6:27:02<1:01:11,  1.80s/it]
                                                         

 86%|████████▌ | 12690/14725 [6:27:02<1:01:11,  1.80s/it]
 86%|████████▌ | 12691/14725 [6:27:04<1:01:22,  1.81s/it]
 86%|████████▌ | 12692/14725 [6:27:06<1:01:59,  1.83s/it]
 86%|████████▌ | 12693/14725 [6:27:07<1:01:41,  1.82s/it]
 86%|████████▌ | 12694/14725 [6:27:09<1:01:33,  1.82s/it]
 86%|████████▌ | 12695/14725 [6:27:11<1:01:21,  1.81s/it]
 86%|████████▌ | 12696/14725 [6:27:13<1:00:41,  1.79s/it]
 86%|████████▌ | 12697/14725 [6:27:14<1:00:17,  1.78s/it]
 86%|████████▌ | 12698/14725 [6:27:16<1:00:42,  1.80s/it]
 86%|████████▌ | 12699/14725 [6:27:1
+0: {'loss': 0.8025, 'grad_norm': 0.6375809358088981, 'learning_rate': 2e-05, 'epoch': 0.86}
+0: 8<1:00:33,  1.79s/it]
 86%|████████▌ | 12700/14725 [6:27:20<1:00:46,  1.80s/it]
                                                         

 86%|████████▌ | 12700/14725 [6:27:20<1:00:46,  1.80s/it]
 86%|████████▋ | 12701/14725 [6:27:22<1:00:43,  1.80s/it]
 86%|████████▋ | 12702/14725 [6:27:23<1:00:56,  1.81s/it]
 86%|████████▋ | 12703/14725 [6:27:25<1:01:41,  1.83s/it]
 86%|████████▋ | 12704/14725 [6:27:27<1:00:57,  1.81s/it]
 86%|████████▋ | 12705/14725 [6:27:29<1:00:34,  1.80s/it]
 86%|████████▋ | 12706/14725 [6:27:31<1:00:56,  1.81s/it]
 86%|████████▋ | 12707/14725 [6:27:33<1:00:43,  1.81s/it]
 86%|████████▋ | 12708/14725 [6:27:34<1:00:20,  1.80s/it]
 86%|████████▋ | 12709/14725 [6:27:36<1:00:35,  1.80s/it]
 86%|████████▋ | 12710/14725 [6:27:38<1:00:26,  1.80s/it]
                               
+0: {'loss': 0.7835, 'grad_norm': 0.6625345690094853, 'learning_rate': 2e-05, 'epoch': 0.86}
+0: {'loss': 0.7968, 'grad_norm': 0.6814704735469105, 'learning_rate': 2e-05, 'epoch': 0.86}
+0:                           

 86%|████████▋ | 12710/14725 [6:27:38<1:00:26,  1.80s/it]
 86%|████████▋ | 12711/14725 [6:27:40<1:00:29,  1.80s/it]
 86%|████████▋ | 12712/14725 [6:27:42<1:00:52,  1.81s/it]
 86%|████████▋ | 12713/14725 [6:27:43<1:00:27,  1.80s/it]
 86%|████████▋ | 12714/14725 [6:27:45<1:00:20,  1.80s/it]
 86%|████████▋ | 12715/14725 [6:27:47<1:00:04,  1.79s/it]
 86%|████████▋ | 12716/14725 [6:27:49<1:00:49,  1.82s/it]
 86%|████████▋ | 12717/14725 [6:27:51<1:00:50,  1.82s/it]
 86%|████████▋ | 12718/14725 [6:27:52<1:00:24,  1.81s/it]
 86%|████████▋ | 12719/14725 [6:27:54<59:44,  1.79s/it]  
 86%|████████▋ | 12720/14725 [6:27:56<59:51,  1.79s/it]
                                                       

 86%|████████▋ | 12720/14725 [6:27:56<59:51,  1.79s/it]
 86%|████████�
+0: {'loss': 0.7836, 'grad_norm': 0.6277507733665267, 'learning_rate': 2e-05, 'epoch': 0.86}
+0: � | 12721/14725 [6:27:58<1:00:32,  1.81s/it]
 86%|████████▋ | 12722/14725 [6:28:00<1:00:22,  1.81s/it]
 86%|████████▋ | 12723/14725 [6:28:01<59:39,  1.79s/it]  
 86%|████████▋ | 12724/14725 [6:28:03<59:46,  1.79s/it]
 86%|████████▋ | 12725/14725 [6:28:05<1:00:13,  1.81s/it]
 86%|████████▋ | 12726/14725 [6:28:07<1:00:31,  1.82s/it]
 86%|████████▋ | 12727/14725 [6:28:09<1:00:31,  1.82s/it]
 86%|████████▋ | 12728/14725 [6:28:10<59:56,  1.80s/it]  
 86%|████████▋ | 12729/14725 [6:28:12<59:45,  1.80s/it]
 86%|████████▋ | 12730/14725 [6:28:14<59:14,  1.78s/it]
                                                       

 86%|████████▋ | 12730/14725 [6:28:14<59:14,  1.78s/it]
 86%|████████▋ | 12731/14725 [6:28:16<59:35,  1.79s/it]
 86%|████████▋ | 12732/14725 [6:28:18<59:44,  1.80s/it]
 86%|█████�
+0: {'loss': 0.7943, 'grad_norm': 0.7439850220246729, 'learning_rate': 2e-05, 'epoch': 0.87}
+0: �██▋ | 12733/14725 [6:28:19<59:38,  1.80s/it]
 86%|████████▋ | 12734/14725 [6:28:21<59:18,  1.79s/it]
 86%|████████▋ | 12735/14725 [6:28:23<59:43,  1.80s/it]
 86%|████████▋ | 12736/14725 [6:28:25<1:00:27,  1.82s/it]
 86%|████████▋ | 12737/14725 [6:28:27<59:57,  1.81s/it]  
 87%|████████▋ | 12738/14725 [6:28:29<1:01:53,  1.87s/it]
 87%|████████▋ | 12739/14725 [6:28:30<1:00:38,  1.83s/it]
 87%|████████▋ | 12740/14725 [6:28:32<59:57,  1.81s/it]  
                                                       

 87%|████████▋ | 12740/14725 [6:28:32<59:57,  1.81s/it]
 87%|████████▋ | 12741/14725 [6:28:34<1:00:01,  1.82s/it]
 87%|████████▋ | 12742/14725 [6:28:36<59:46,  1.81s/it]  
 87%|████████▋ | 12743/14725 [6:28:38<59:16,  1.79s/it]
 87%|████████▋ | 12744/14725 [6:28:39<59:38,  1.81s/it]
 87%|██�
+0: {'loss': 0.7959, 'grad_norm': 0.7054969779443883, 'learning_rate': 2e-05, 'epoch': 0.87}
+0: �█████▋ | 12745/14725 [6:28:41<59:22,  1.80s/it]
 87%|████████▋ | 12746/14725 [6:28:43<59:17,  1.80s/it]
 87%|████████▋ | 12747/14725 [6:28:45<59:35,  1.81s/it]
 87%|████████▋ | 12748/14725 [6:28:47<59:21,  1.80s/it]
 87%|████████▋ | 12749/14725 [6:28:48<59:10,  1.80s/it]
 87%|████████▋ | 12750/14725 [6:28:50<59:27,  1.81s/it]
                                                       

 87%|████████▋ | 12750/14725 [6:28:50<59:27,  1.81s/it]
 87%|████████▋ | 12751/14725 [6:28:52<59:33,  1.81s/it]
 87%|████████▋ | 12752/14725 [6:28:54<58:57,  1.79s/it]
 87%|████████▋ | 12753/14725 [6:28:56<59:51,  1.82s/it]
 87%|████████▋ | 12754/14725 [6:28:57<1:00:07,  1.83s/it]
 87%|████████▋ | 12755/14725 [6:28:59<1:00:25,  1.84s/it]
 87%|████████▋ | 12756/14725 [6:29:01<59:44,  1.82s/it]  
 87%|██�
+0: {'loss': 0.8033, 'grad_norm': 0.6495391666003058, 'learning_rate': 2e-05, 'epoch': 0.87}
+0: ��█████▋ | 12757/14725 [6:29:03<59:17,  1.81s/it]
 87%|████████▋ | 12758/14725 [6:29:05<59:16,  1.81s/it]
 87%|████████▋ | 12759/14725 [6:29:06<59:04,  1.80s/it]
 87%|████████▋ | 12760/14725 [6:29:08<59:05,  1.80s/it]
                                                       

 87%|████████▋ | 12760/14725 [6:29:08<59:05,  1.80s/it]
 87%|████████▋ | 12761/14725 [6:29:10<58:50,  1.80s/it]
 87%|████████▋ | 12762/14725 [6:29:12<58:26,  1.79s/it]
 87%|████████▋ | 12763/14725 [6:29:14<58:14,  1.78s/it]
 87%|████████▋ | 12764/14725 [6:29:15<57:56,  1.77s/it]
 87%|████████▋ | 12765/14725 [6:29:17<58:01,  1.78s/it]
 87%|████████▋ | 12766/14725 [6:29:19<58:31,  1.79s/it]
 87%|████████▋ | 12767/14725 [6:29:21<58:33,  1.79s/it]
 87%|████████▋ | 12768/14725 [6:29:23<58:52,  1.80s/it]
 87%|████
+0: {'loss': 0.8059, 'grad_norm': 0.6720900859314374, 'learning_rate': 2e-05, 'epoch': 0.87}
+0: ████▋ | 12769/14725 [6:29:24<59:03,  1.81s/it]
 87%|████████▋ | 12770/14725 [6:29:26<59:15,  1.82s/it]
                                                       

 87%|████████▋ | 12770/14725 [6:29:26<59:15,  1.82s/it]
 87%|████████▋ | 12771/14725 [6:29:28<59:00,  1.81s/it]
 87%|████████▋ | 12772/14725 [6:29:30<59:14,  1.82s/it]
 87%|████████▋ | 12773/14725 [6:29:32<58:50,  1.81s/it]
 87%|████████▋ | 12774/14725 [6:29:33<58:39,  1.80s/it]
 87%|████████▋ | 12775/14725 [6:29:35<58:20,  1.79s/it]
 87%|████████▋ | 12776/14725 [6:29:37<58:08,  1.79s/it]
 87%|████████▋ | 12777/14725 [6:29:39<59:47,  1.84s/it]
 87%|████████▋ | 12778/14725 [6:29:41<59:04,  1.82s/it]
 87%|████████▋ | 12779/14725 [6:29:43<58:40,  1.81s/it]
 87%|████████▋ | 12780/14725 [6:29:44<58:25,  1.80s/it]
                      
+0: {'loss': 0.806, 'grad_norm': 0.6680400870676796, 'learning_rate': 2e-05, 'epoch': 0.87}
+0: {'loss': 0.7818, 'grad_norm': 0.6286159928299893, 'learning_rate': 2e-05, 'epoch': 0.87}
+0:                                  

 87%|████████▋ | 12780/14725 [6:29:44<58:25,  1.80s/it]
 87%|████████▋ | 12781/14725 [6:29:46<58:47,  1.81s/it]
 87%|████████▋ | 12782/14725 [6:29:48<58:40,  1.81s/it]
 87%|████████▋ | 12783/14725 [6:29:50<59:17,  1.83s/it]
 87%|████████▋ | 12784/14725 [6:29:52<58:46,  1.82s/it]
 87%|████████▋ | 12785/14725 [6:29:53<59:07,  1.83s/it]
 87%|████████▋ | 12786/14725 [6:29:55<58:39,  1.82s/it]
 87%|████████▋ | 12787/14725 [6:29:57<58:12,  1.80s/it]
 87%|████████▋ | 12788/14725 [6:29:59<58:37,  1.82s/it]
 87%|████████▋ | 12789/14725 [6:30:01<58:10,  1.80s/it]
 87%|████████▋ | 12790/14725 [6:30:02<58:03,  1.80s/it]
                                                       

 87%|████████▋ | 12790/14725 [6:30:02<58:03,  1.80s/it]
 87%|████████▋ | 12791/147
+0: {'loss': 0.8035, 'grad_norm': 0.6666606312888089, 'learning_rate': 2e-05, 'epoch': 0.87}
+0: 25 [6:30:04<59:06,  1.83s/it]
 87%|████████▋ | 12792/14725 [6:30:06<58:29,  1.82s/it]
 87%|████████▋ | 12793/14725 [6:30:08<58:10,  1.81s/it]
 87%|████████▋ | 12794/14725 [6:30:10<58:44,  1.83s/it]
 87%|████████▋ | 12795/14725 [6:30:12<58:29,  1.82s/it]
 87%|████████▋ | 12796/14725 [6:30:13<58:26,  1.82s/it]
 87%|████████▋ | 12797/14725 [6:30:15<58:06,  1.81s/it]
 87%|████████▋ | 12798/14725 [6:30:17<58:04,  1.81s/it]
 87%|████████▋ | 12799/14725 [6:30:19<58:26,  1.82s/it]
 87%|████████▋ | 12800/14725 [6:30:21<58:44,  1.83s/it]
                                                       

 87%|████████▋ | 12800/14725 [6:30:21<58:44,  1.83s/it]
 87%|████████▋ | 12801/14725 [6:30:23<58:28,  1.82s/it]
 87%|████████▋ | 12802/14725 [6:30:24<57:40,  1.80s/it]
 87%|████████▋ | 12803/14725 [6
+0: {'loss': 0.7863, 'grad_norm': 0.6487162671098746, 'learning_rate': 2e-05, 'epoch': 0.87}
+0: :30:26<57:38,  1.80s/it]
 87%|████████▋ | 12804/14725 [6:30:28<57:52,  1.81s/it]
 87%|████████▋ | 12805/14725 [6:30:30<58:37,  1.83s/it]
 87%|████████▋ | 12806/14725 [6:30:32<58:46,  1.84s/it]
 87%|████████▋ | 12807/14725 [6:30:33<58:20,  1.83s/it]
 87%|████████▋ | 12808/14725 [6:30:35<57:34,  1.80s/it]
 87%|████████▋ | 12809/14725 [6:30:37<57:22,  1.80s/it]
 87%|████████▋ | 12810/14725 [6:30:39<57:25,  1.80s/it]
                                                       

 87%|████████▋ | 12810/14725 [6:30:39<57:25,  1.80s/it]
 87%|████████▋ | 12811/14725 [6:30:41<57:16,  1.80s/it]
 87%|████████▋ | 12812/14725 [6:30:42<56:51,  1.78s/it]
 87%|████████▋ | 12813/14725 [6:30:44<57:00,  1.79s/it]
 87%|████████▋ | 12814/14725 [6:30:46<57:30,  1.81s/it]
 87%|████████▋ | 12815/14725 [6:30:4
+0: {'loss': 0.777, 'grad_norm': 0.7013349611830145, 'learning_rate': 2e-05, 'epoch': 0.87}
+0: 8<57:21,  1.80s/it]
 87%|████████▋ | 12816/14725 [6:30:50<57:21,  1.80s/it]
 87%|████████▋ | 12817/14725 [6:30:51<57:29,  1.81s/it]
 87%|████████▋ | 12818/14725 [6:30:53<57:36,  1.81s/it]
 87%|████████▋ | 12819/14725 [6:30:55<57:03,  1.80s/it]
 87%|████████▋ | 12820/14725 [6:30:57<57:45,  1.82s/it]
                                                       

 87%|████████▋ | 12820/14725 [6:30:57<57:45,  1.82s/it]
 87%|████████▋ | 12821/14725 [6:30:59<57:13,  1.80s/it]
 87%|████████▋ | 12822/14725 [6:31:01<59:04,  1.86s/it]
 87%|████████▋ | 12823/14725 [6:31:02<58:24,  1.84s/it]
 87%|████████▋ | 12824/14725 [6:31:04<57:34,  1.82s/it]
 87%|████████▋ | 12825/14725 [6:31:06<58:10,  1.84s/it]
 87%|████████▋ | 12826/14725 [6:31:08<58:57,  1.86s/it]
 87%|████████▋ | 12827/14725 [6:31:10<58:
+0: {'loss': 0.804, 'grad_norm': 0.6956335320414042, 'learning_rate': 2e-05, 'epoch': 0.87}
+0: 38,  1.85s/it]
 87%|████████▋ | 12828/14725 [6:31:12<58:00,  1.83s/it]
 87%|████████▋ | 12829/14725 [6:31:13<57:45,  1.83s/it]
 87%|████████▋ | 12830/14725 [6:31:15<58:02,  1.84s/it]
                                                       

 87%|████████▋ | 12830/14725 [6:31:15<58:02,  1.84s/it]
 87%|████████▋ | 12831/14725 [6:31:17<58:24,  1.85s/it]
 87%|████████▋ | 12832/14725 [6:31:19<58:12,  1.84s/it]
 87%|████████▋ | 12833/14725 [6:31:21<58:25,  1.85s/it]
 87%|████████▋ | 12834/14725 [6:31:23<57:27,  1.82s/it]
 87%|████████▋ | 12835/14725 [6:31:24<56:59,  1.81s/it]
 87%|████████▋ | 12836/14725 [6:31:26<59:09,  1.88s/it]
 87%|████████▋ | 12837/14725 [6:31:28<58:18,  1.85s/it]
 87%|████████▋ | 12838/14725 [6:31:30<57:33,  1.83s/it]
 87%|████████▋ | 12839/14725 [6:31:32<57:35,  
+0: {'loss': 0.7849, 'grad_norm': 0.7065184954507657, 'learning_rate': 2e-05, 'epoch': 0.87}
+0: {'loss': 0.8118, 'grad_norm': 0.6405600544539416, 'learning_rate': 2e-05, 'epoch': 0.87}
+0: 1.83s/it]
 87%|████████▋ | 12840/14725 [6:31:34<58:59,  1.88s/it]
                                                       

 87%|████████▋ | 12840/14725 [6:31:34<58:59,  1.88s/it]
 87%|████████▋ | 12841/14725 [6:31:36<58:23,  1.86s/it]
 87%|████████▋ | 12842/14725 [6:31:37<57:46,  1.84s/it]
 87%|████████▋ | 12843/14725 [6:31:39<58:44,  1.87s/it]
 87%|████████▋ | 12844/14725 [6:31:41<58:12,  1.86s/it]
 87%|████████▋ | 12845/14725 [6:31:43<59:59,  1.91s/it]
 87%|████████▋ | 12846/14725 [6:31:45<59:27,  1.90s/it]
 87%|████████▋ | 12847/14725 [6:31:47<58:59,  1.88s/it]
 87%|████████▋ | 12848/14725 [6:31:49<58:57,  1.88s/it]
 87%|████████▋ | 12849/14725 [6:31:51<58:54,  1.88s/it]
 87%|████████▋ | 12850/14725 [6:31:53<58:13,  1.86s/it]
                                                       

 87%|██�
+0: {'loss': 0.7844, 'grad_norm': 0.6727627778753659, 'learning_rate': 2e-05, 'epoch': 0.87}
+0: ��█████▋ | 12850/14725 [6:31:53<58:13,  1.86s/it]
 87%|████████▋ | 12851/14725 [6:31:54<57:30,  1.84s/it]
 87%|████████▋ | 12852/14725 [6:31:56<57:45,  1.85s/it]
 87%|████████▋ | 12853/14725 [6:31:58<57:48,  1.85s/it]
 87%|████████▋ | 12854/14725 [6:32:00<56:48,  1.82s/it]
 87%|████████▋ | 12855/14725 [6:32:02<56:14,  1.80s/it]
 87%|████████▋ | 12856/14725 [6:32:03<56:05,  1.80s/it]
 87%|████████▋ | 12857/14725 [6:32:05<55:33,  1.78s/it]
 87%|████████▋ | 12858/14725 [6:32:07<55:49,  1.79s/it]
 87%|████████▋ | 12859/14725 [6:32:09<55:59,  1.80s/it]
 87%|████████▋ | 12860/14725 [6:32:10<55:30,  1.79s/it]
                                                       

 87%|████████▋ | 12860/14725 [6:32:10<55:30,  1.79s/it]
 87%|████████▋ | 12861/14725 [6:32:12<55:43,  1.79s/it]
 87%|████
+0: {'loss': 0.8051, 'grad_norm': 0.672773101927267, 'learning_rate': 2e-05, 'epoch': 0.87}
+0: ████▋ | 12862/14725 [6:32:14<55:41,  1.79s/it]
 87%|████████▋ | 12863/14725 [6:32:16<55:45,  1.80s/it]
 87%|████████▋ | 12864/14725 [6:32:18<55:31,  1.79s/it]
 87%|████████▋ | 12865/14725 [6:32:19<55:52,  1.80s/it]
 87%|████████▋ | 12866/14725 [6:32:21<56:27,  1.82s/it]
 87%|████████▋ | 12867/14725 [6:32:23<56:30,  1.83s/it]
 87%|████████▋ | 12868/14725 [6:32:25<55:59,  1.81s/it]
 87%|████████▋ | 12869/14725 [6:32:27<55:25,  1.79s/it]
 87%|████████▋ | 12870/14725 [6:32:29<55:27,  1.79s/it]
                                                       

 87%|████████▋ | 12870/14725 [6:32:29<55:27,  1.79s/it]
 87%|████████▋ | 12871/14725 [6:32:30<55:26,  1.79s/it]
 87%|████████▋ | 12872/14725 [6:32:32<55:37,  1.80s/it]
 87%|████████▋ | 12873/14725 [6:32:34<55:09,  1.79s/it]
 87%|█████�
+0: {'loss': 0.7957, 'grad_norm': 0.6510758108789501, 'learning_rate': 2e-05, 'epoch': 0.87}
+0: �██▋ | 12874/14725 [6:32:36<54:54,  1.78s/it]
 87%|████████▋ | 12875/14725 [6:32:37<54:53,  1.78s/it]
 87%|████████▋ | 12876/14725 [6:32:39<54:58,  1.78s/it]
 87%|████████▋ | 12877/14725 [6:32:41<54:48,  1.78s/it]
 87%|████████▋ | 12878/14725 [6:32:43<55:19,  1.80s/it]
 87%|████████▋ | 12879/14725 [6:32:45<55:08,  1.79s/it]
 87%|████████▋ | 12880/14725 [6:32:46<54:53,  1.79s/it]
                                                       

 87%|████████▋ | 12880/14725 [6:32:46<54:53,  1.79s/it]
 87%|████████▋ | 12881/14725 [6:32:48<55:22,  1.80s/it]
 87%|████████▋ | 12882/14725 [6:32:50<55:31,  1.81s/it]
 87%|████████▋ | 12883/14725 [6:32:52<55:35,  1.81s/it]
 87%|████████▋ | 12884/14725 [6:32:54<55:54,  1.82s/it]
 88%|████████▊ | 12885/14725 [6:32:56<55:53,  1.82s/it]
 88%|███████�
+0: {'loss': 0.8026, 'grad_norm': 0.6632489802478302, 'learning_rate': 2e-05, 'epoch': 0.88}
+0: ��▊ | 12886/14725 [6:32:57<55:31,  1.81s/it]
 88%|████████▊ | 12887/14725 [6:32:59<55:37,  1.82s/it]
 88%|████████▊ | 12888/14725 [6:33:01<55:14,  1.80s/it]
 88%|████████▊ | 12889/14725 [6:33:03<54:43,  1.79s/it]
 88%|████████▊ | 12890/14725 [6:33:04<54:22,  1.78s/it]
                                                       

 88%|████████▊ | 12890/14725 [6:33:04<54:22,  1.78s/it]
 88%|████████▊ | 12891/14725 [6:33:06<54:15,  1.78s/it]
 88%|████████▊ | 12892/14725 [6:33:08<55:30,  1.82s/it]
 88%|████████▊ | 12893/14725 [6:33:10<55:18,  1.81s/it]
 88%|████████▊ | 12894/14725 [6:33:12<55:02,  1.80s/it]
 88%|████████▊ | 12895/14725 [6:33:14<56:13,  1.84s/it]
 88%|████████▊ | 12896/14725 [6:33:15<55:27,  1.82s/it]
 88%|████████▊ | 12897/14725 [6:33:17<55:35,  1.82s/it]
 88%|████████▊
+0: {'loss': 0.8004, 'grad_norm': 0.7072152543524227, 'learning_rate': 2e-05, 'epoch': 0.88}
+0:  | 12898/14725 [6:33:19<55:01,  1.81s/it]
 88%|████████▊ | 12899/14725 [6:33:21<54:42,  1.80s/it]
 88%|████████▊ | 12900/14725 [6:33:23<54:51,  1.80s/it]
                                                       

 88%|████████▊ | 12900/14725 [6:33:23<54:51,  1.80s/it]
 88%|████████▊ | 12901/14725 [6:33:24<55:25,  1.82s/it]
 88%|████████▊ | 12902/14725 [6:33:26<54:58,  1.81s/it]
 88%|████████▊ | 12903/14725 [6:33:28<54:33,  1.80s/it]
 88%|████████▊ | 12904/14725 [6:33:30<54:43,  1.80s/it]
 88%|████████▊ | 12905/14725 [6:33:32<54:39,  1.80s/it]
 88%|████████▊ | 12906/14725 [6:33:33<54:41,  1.80s/it]
 88%|████████▊ | 12907/14725 [6:33:35<54:41,  1.80s/it]
 88%|████████▊ | 12908/14725 [6:33:37<54:52,  1.81s/it]
 88%|████████▊ | 12909/14725 [6:33:39<55:09,  1.82s/it]
 88%|████████▊ | 12
+0: {'loss': 0.8069, 'grad_norm': 0.6888306898932273, 'learning_rate': 2e-05, 'epoch': 0.88}
+0: {'loss': 0.7836, 'grad_norm': 0.6663531824266757, 'learning_rate': 2e-05, 'epoch': 0.88}
+0: 910/14725 [6:33:41<55:06,  1.82s/it]
                                                       

 88%|████████▊ | 12910/14725 [6:33:41<55:06,  1.82s/it]
 88%|████████▊ | 12911/14725 [6:33:43<54:49,  1.81s/it]
 88%|████████▊ | 12912/14725 [6:33:44<54:35,  1.81s/it]
 88%|████████▊ | 12913/14725 [6:33:46<54:15,  1.80s/it]
 88%|████████▊ | 12914/14725 [6:33:48<54:57,  1.82s/it]
 88%|████████▊ | 12915/14725 [6:33:50<54:51,  1.82s/it]
 88%|████████▊ | 12916/14725 [6:33:52<54:32,  1.81s/it]
 88%|████████▊ | 12917/14725 [6:33:53<54:57,  1.82s/it]
 88%|████████▊ | 12918/14725 [6:33:55<54:24,  1.81s/it]
 88%|████████▊ | 12919/14725 [6:33:57<53:54,  1.79s/it]
 88%|████████▊ | 12920/14725 [6:33:59<53:44,  1.79s/it]
                                                       

 88%|████████▊ | 12920/14725 [6:33:59<53:
+0: {'loss': 0.7933, 'grad_norm': 0.694996992423342, 'learning_rate': 2e-05, 'epoch': 0.88}
+0: 44,  1.79s/it]
 88%|████████▊ | 12921/14725 [6:34:00<53:35,  1.78s/it]
 88%|████████▊ | 12922/14725 [6:34:02<54:21,  1.81s/it]
 88%|████████▊ | 12923/14725 [6:34:04<53:46,  1.79s/it]
 88%|████████▊ | 12924/14725 [6:34:06<54:22,  1.81s/it]
 88%|████████▊ | 12925/14725 [6:34:08<53:48,  1.79s/it]
 88%|████████▊ | 12926/14725 [6:34:09<53:22,  1.78s/it]
 88%|████████▊ | 12927/14725 [6:34:11<53:00,  1.77s/it]
 88%|████████▊ | 12928/14725 [6:34:13<53:32,  1.79s/it]
 88%|████████▊ | 12929/14725 [6:34:15<53:32,  1.79s/it]
 88%|████████▊ | 12930/14725 [6:34:17<54:00,  1.81s/it]
                                                       

 88%|████████▊ | 12930/14725 [6:34:17<54:00,  1.81s/it]
 88%|████████▊ | 12931/14725 [6:34:18<54:00,  1.81s/it]
 88%|████████▊ | 12932/14725 [6:34:20<53:25,  
+0: {'loss': 0.7675, 'grad_norm': 0.6515856903518645, 'learning_rate': 2e-05, 'epoch': 0.88}
+0: 1.79s/it]
 88%|████████▊ | 12933/14725 [6:34:22<53:28,  1.79s/it]
 88%|████████▊ | 12934/14725 [6:34:24<53:24,  1.79s/it]
 88%|████████▊ | 12935/14725 [6:34:26<53:21,  1.79s/it]
 88%|████████▊ | 12936/14725 [6:34:27<53:25,  1.79s/it]
 88%|████████▊ | 12937/14725 [6:34:29<54:03,  1.81s/it]
 88%|████████▊ | 12938/14725 [6:34:31<53:32,  1.80s/it]
 88%|████████▊ | 12939/14725 [6:34:33<53:27,  1.80s/it]
 88%|████████▊ | 12940/14725 [6:34:35<53:29,  1.80s/it]
                                                       

 88%|████████▊ | 12940/14725 [6:34:35<53:29,  1.80s/it]
 88%|████████▊ | 12941/14725 [6:34:36<53:31,  1.80s/it]
 88%|████████▊ | 12942/14725 [6:34:38<54:01,  1.82s/it]
 88%|████████▊ | 12943/14725 [6:34:40<53:52,  1.81s/it]
 88%|████████▊ | 12944/14725 [6:34:42<53:42,  1.81s
+0: {'loss': 0.7966, 'grad_norm': 0.6953874918450177, 'learning_rate': 2e-05, 'epoch': 0.88}
+0: /it]
 88%|████████▊ | 12945/14725 [6:34:44<54:11,  1.83s/it]
 88%|███���████▊ | 12946/14725 [6:34:46<53:58,  1.82s/it]
 88%|████████▊ | 12947/14725 [6:34:47<53:28,  1.80s/it]
 88%|████████▊ | 12948/14725 [6:34:49<53:08,  1.79s/it]
 88%|████████▊ | 12949/14725 [6:34:51<53:23,  1.80s/it]
 88%|████████▊ | 12950/14725 [6:34:53<53:16,  1.80s/it]
                                                       

 88%|████████▊ | 12950/14725 [6:34:53<53:16,  1.80s/it]
 88%|████████▊ | 12951/14725 [6:34:55<53:05,  1.80s/it]
 88%|████████▊ | 12952/14725 [6:34:56<53:21,  1.81s/it]
 88%|████████▊ | 12953/14725 [6:34:58<53:25,  1.81s/it]
 88%|████████▊ | 12954/14725 [6:35:00<53:13,  1.80s/it]
 88%|████████▊ | 12955/14725 [6:35:02<52:41,  1.79s/it]
 88%|████████▊ | 12956/14725 [6:35:03<52:43,  1.79s/it]
+0: {'loss': 0.7997, 'grad_norm': 0.6890823781254372, 'learning_rate': 2e-05, 'epoch': 0.88}
+0:  88%|████████▊ | 12957/14725 [6:35:05<53:27,  1.81s/it]
 88%|████████▊ | 12958/14725 [6:35:07<53:27,  1.82s/it]
 88%|████████▊ | 12959/14725 [6:35:09<52:59,  1.80s/it]
 88%|████████▊ | 12960/14725 [6:35:11<54:10,  1.84s/it]
                                                       

 88%|████████▊ | 12960/14725 [6:35:11<54:10,  1.84s/it]
 88%|████████▊ | 12961/14725 [6:35:13<53:18,  1.81s/it]
 88%|████████▊ | 12962/14725 [6:35:14<53:13,  1.81s/it]
 88%|████████▊ | 12963/14725 [6:35:16<53:08,  1.81s/it]
 88%|████████▊ | 12964/14725 [6:35:18<53:06,  1.81s/it]
 88%|████████▊ | 12965/14725 [6:35:20<52:37,  1.79s/it]
 88%|████████▊ | 12966/14725 [6:35:22<53:16,  1.82s/it]
 88%|████████▊ | 12967/14725 [6:35:23<53:05,  1.81s/it]
 88%|████████▊ | 12968/14725 [6:35:25<52:54,  1.81s/it]
 88%|
+0: {'loss': 0.7953, 'grad_norm': 0.6902312180674318, 'learning_rate': 2e-05, 'epoch': 0.88}
+0: ████████▊ | 12969/14725 [6:35:27<52:27,  1.79s/it]
 88%|████████▊ | 12970/14725 [6:35:29<52:36,  1.80s/it]
                                                       

 88%|████████▊ | 12970/14725 [6:35:29<52:36,  1.80s/it]
 88%|████████▊ | 12971/14725 [6:35:31<52:20,  1.79s/it]
 88%|████████▊ | 12972/14725 [6:35:32<52:10,  1.79s/it]
 88%|████████▊ | 12973/14725 [6:35:34<52:47,  1.81s/it]
 88%|████████▊ | 12974/14725 [6:35:36<52:10,  1.79s/it]
 88%|████████▊ | 12975/14725 [6:35:38<52:00,  1.78s/it]
 88%|████████▊ | 12976/14725 [6:35:40<52:45,  1.81s/it]
 88%|████████▊ | 12977/14725 [6:35:42<53:18,  1.83s/it]
 88%|████████▊ | 12978/14725 [6:35:43<53:13,  1.83s/it]
 88%|████████▊ | 12979/14725 [6:35:45<52:34,  1.81s/it]
 88%|████████▊ | 12980/14725 [6:35:47<52:17,  1.80s/it]
          
+0: {'loss': 0.7726, 'grad_norm': 0.6558820954186129, 'learning_rate': 2e-05, 'epoch': 0.88}
+0: {'loss': 0.781, 'grad_norm': 0.6502677859776366, 'learning_rate': 2e-05, 'epoch': 0.88}
+0:                                              

 88%|████████▊ | 12980/14725 [6:35:47<52:17,  1.80s/it]
 88%|████████▊ | 12981/14725 [6:35:49<52:45,  1.82s/it]
 88%|████████▊ | 12982/14725 [6:35:51<53:26,  1.84s/it]
 88%|████████▊ | 12983/14725 [6:35:52<52:42,  1.82s/it]
 88%|████████▊ | 12984/14725 [6:35:54<52:12,  1.80s/it]
 88%|████████▊ | 12985/14725 [6:35:56<52:01,  1.79s/it]
 88%|████████▊ | 12986/14725 [6:35:58<51:47,  1.79s/it]
 88%|████████▊ | 12987/14725 [6:35:59<51:44,  1.79s/it]
 88%|████████▊ | 12988/14725 [6:36:01<52:21,  1.81s/it]
 88%|████████▊ | 12989/14725 [6:36:03<51:58,  1.80s/it]
 88%|████████▊ | 12990/14725 [6:36:05<51:29,  1.78s/it]
                                                       

 88%|████████▊ | 12990/14725 [6:36:05<51:29,  1.78s/it]
 88%|████████▊
+0: {'loss': 0.8147, 'grad_norm': 0.6657170972610306, 'learning_rate': 2e-05, 'epoch': 0.88}
+0:  | 12991/14725 [6:36:07<52:03,  1.80s/it]
 88%|████████▊ | 12992/14725 [6:36:09<52:55,  1.83s/it]
 88%|████████▊ | 12993/14725 [6:36:10<52:17,  1.81s/it]
 88%|████████▊ | 12994/14725 [6:36:12<52:32,  1.82s/it]
 88%|████████▊ | 12995/14725 [6:36:14<52:10,  1.81s/it]
 88%|████████▊ | 12996/14725 [6:36:16<51:51,  1.80s/it]
 88%|████████▊ | 12997/14725 [6:36:18<51:26,  1.79s/it]
 88%|████████▊ | 12998/14725 [6:36:19<52:00,  1.81s/it]
 88%|████████▊ | 12999/14725 [6:36:21<52:06,  1.81s/it]
 88%|████████▊ | 13000/14725 [6:36:23<51:49,  1.80s/it]
                                                       

 88%|████████▊ | 13000/14725 [6:36:23<51:49,  1.80s/it]
 88%|████████▊ | 13001/14725 [6:36:25<51:48,  1.80s/it]
 88%|████████▊ | 13002/14725 [6:36:27<51:27,  1.79s/it]
 88%|████████▊ | 13
+0: {'loss': 0.7984, 'grad_norm': 0.6625971093483076, 'learning_rate': 2e-05, 'epoch': 0.88}
+0: 003/14725 [6:36:28<51:12,  1.78s/it]
 88%|████████▊ | 13004/14725 [6:36:30<51:18,  1.79s/it]
 88%|████████▊ | 13005/14725 [6:36:32<51:11,  1.79s/it]
 88%|████████▊ | 13006/14725 [6:36:34<51:51,  1.81s/it]
 88%|████████▊ | 13007/14725 [6:36:36<51:35,  1.80s/it]
 88%|████████▊ | 13008/14725 [6:36:37<51:54,  1.81s/it]
 88%|████████▊ | 13009/14725 [6:36:39<51:21,  1.80s/it]
 88%|████████▊ | 13010/14725 [6:36:41<51:15,  1.79s/it]
                                                       

 88%|████████▊ | 13010/14725 [6:36:41<51:15,  1.79s/it]
 88%|████████▊ | 13011/14725 [6:36:43<51:11,  1.79s/it]
 88%|████████▊ | 13012/14725 [6:36:45<51:29,  1.80s/it]
 88%|████████▊ | 13013/14725 [6:36:46<51:45,  1.81s/it]
 88%|████████▊ | 13014/14725 [6:36:48<51:42,  1.81s/it]
 88%|████████▊ | 13015/1
+0: {'loss': 0.7865, 'grad_norm': 0.7165441735630073, 'learning_rate': 2e-05, 'epoch': 0.88}
+0: 4725 [6:36:50<51:24,  1.80s/it]
 88%|████████▊ | 13016/14725 [6:36:52<51:20,  1.80s/it]
 88%|████████▊ | 13017/14725 [6:36:54<50:59,  1.79s/it]
 88%|████████▊ | 13018/14725 [6:36:55<50:45,  1.78s/it]
 88%|████████▊ | 13019/14725 [6:36:57<51:32,  1.81s/it]
 88%|████████▊ | 13020/14725 [6:36:59<51:39,  1.82s/it]
                                                       

 88%|████████▊ | 13020/14725 [6:36:59<51:39,  1.82s/it]
 88%|████████▊ | 13021/14725 [6:37:01<51:59,  1.83s/it]
 88%|████████▊ | 13022/14725 [6:37:03<51:47,  1.82s/it]
 88%|████████▊ | 13023/14725 [6:37:04<51:25,  1.81s/it]
 88%|████████▊ | 13024/14725 [6:37:06<51:08,  1.80s/it]
 88%|████████▊ | 13025/14725 [6:37:08<52:56,  1.87s/it]
 88%|████████▊ | 13026/14725 [6:37:10<52:17,  1.85s/it]
 88%|████████▊ | 13027/14725 
+0: {'loss': 0.8005, 'grad_norm': 0.7437114498227428, 'learning_rate': 2e-05, 'epoch': 0.88}
+0: [6:37:12<51:39,  1.83s/it]
 88%|████████▊ | 13028/14725 [6:37:14<51:08,  1.81s/it]
 88%|████████▊ | 13029/14725 [6:37:15<50:37,  1.79s/it]
 88%|████████▊ | 13030/14725 [6:37:17<51:33,  1.83s/it]
                                                       

 88%|████████▊ | 13030/14725 [6:37:17<51:33,  1.83s/it]
 88%|████████▊ | 13031/14725 [6:37:19<51:07,  1.81s/it]
 89%|████████▊ | 13032/14725 [6:37:21<50:51,  1.80s/it]
 89%|████████▊ | 13033/14725 [6:37:23<50:24,  1.79s/it]
 89%|████████▊ | 13034/14725 [6:37:24<50:10,  1.78s/it]
 89%|████████▊ | 13035/14725 [6:37:26<50:11,  1.78s/it]
 89%|████████▊ | 13036/14725 [6:37:28<50:25,  1.79s/it]
 89%|████████▊ | 13037/14725 [6:37:30<50:26,  1.79s/it]
 89%|████████▊ | 13038/14725 [6:37:32<51:00,  1.81s/it]
 89%|████████▊ | 13039/14725 [6:37
+0: {'loss': 0.7863, 'grad_norm': 0.654854765405294, 'learning_rate': 2e-05, 'epoch': 0.89}
+0: {'loss': 0.8068, 'grad_norm': 0.6843885076562619, 'learning_rate': 2e-05, 'epoch': 0.89}
+0: :33<50:34,  1.80s/it]
 89%|████████▊ | 13040/14725 [6:37:35<50:12,  1.79s/it]
                                                       

 89%|████████▊ | 13040/14725 [6:37:35<50:12,  1.79s/it]
 89%|████████▊ | 13041/14725 [6:37:37<51:17,  1.83s/it]
 89%|████████▊ | 13042/14725 [6:37:39<54:41,  1.95s/it]
 89%|████████▊ | 13043/14725 [6:37:41<53:32,  1.91s/it]
 89%|████████▊ | 13044/14725 [6:37:43<52:48,  1.89s/it]
 89%|████████▊ | 13045/14725 [6:37:45<52:42,  1.88s/it]
 89%|████████▊ | 13046/14725 [6:37:47<51:29,  1.84s/it]
 89%|████████▊ | 13047/14725 [6:37:48<51:25,  1.84s/it]
 89%|████████▊ | 13048/14725 [6:37:50<50:58,  1.82s/it]
 89%|████████▊ | 13049/14725 [6:37:52<50:33,  1.81s/it]
 89%|████████▊ | 13050/14725 [6:37:54<50:31,  1.81s/it]
                                                       

+0: {'loss': 0.7881, 'grad_norm': 0.6601912509686955, 'learning_rate': 2e-05, 'epoch': 0.89}
+0:  89%|████████▊ | 13050/14725 [6:37:54<50:31,  1.81s/it]
 89%|████████▊ | 13051/14725 [6:37:56<50:44,  1.82s/it]
 89%|████████▊ | 13052/14725 [6:37:57<50:12,  1.80s/it]
 89%|████████▊ | 13053/14725 [6:37:59<50:59,  1.83s/it]
 89%|████████▊ | 13054/14725 [6:38:01<50:52,  1.83s/it]
 89%|████████▊ | 13055/14725 [6:38:03<53:26,  1.92s/it]
 89%|████████▊ | 13056/14725 [6:38:05<52:14,  1.88s/it]
 89%|████████▊ | 13057/14725 [6:38:07<52:20,  1.88s/it]
 89%|████████▊ | 13058/14725 [6:38:09<51:59,  1.87s/it]
 89%|████████▊ | 13059/14725 [6:38:11<52:14,  1.88s/it]
 89%|████████▊ | 13060/14725 [6:38:12<51:27,  1.85s/it]
                                                       

 89%|████████▊ | 13060/14725 [6:38:12<51:27,  1.85s/it]
 89%|████████▊ | 13061/14725 [6:38:14<50:57,  1.84s/it]
 89%|
+0: {'loss': 0.8166, 'grad_norm': 0.6795269225219914, 'learning_rate': 2e-05, 'epoch': 0.89}
+0: ████████▊ | 13062/14725 [6:38:16<50:44,  1.83s/it]
 89%|████████▊ | 13063/14725 [6:38:18<50:51,  1.84s/it]
 89%|████████▊ | 13064/14725 [6:38:20<51:18,  1.85s/it]
 89%|████████▊ | 13065/14725 [6:38:22<50:18,  1.82s/it]
 89%|████████▊ | 13066/14725 [6:38:23<50:34,  1.83s/it]
 89%|████████▊ | 13067/14725 [6:38:25<50:29,  1.83s/it]
 89%|████████▊ | 13068/14725 [6:38:27<50:13,  1.82s/it]
 89%|████████▉ | 13069/14725 [6:38:29<50:21,  1.82s/it]
 89%|████████▉ | 13070/14725 [6:38:31<50:05,  1.82s/it]
                                                       

 89%|████████▉ | 13070/14725 [6:38:31<50:05,  1.82s/it]
 89%|████████▉ | 13071/14725 [6:38:32<49:42,  1.80s/it]
 89%|████████▉ | 13072/14725 [6:38:34<49:12,  1.79s/it]
 89%|████████▉ | 13073/14725 [6:38:36<49:25,  1.80s/it]
 89%|█�
+0: {'loss': 0.8024, 'grad_norm': 0.667628086689151, 'learning_rate': 2e-05, 'epoch': 0.89}
+0: �██████▉ | 13074/14725 [6:38:38<49:11,  1.79s/it]
 89%|████████▉ | 13075/14725 [6:38:39<48:52,  1.78s/it]
 89%|████████▉ | 13076/14725 [6:38:41<49:01,  1.78s/it]
 89%|████████▉ | 13077/14725 [6:38:43<52:18,  1.90s/it]
 89%|████████▉ | 13078/14725 [6:38:45<51:46,  1.89s/it]
 89%|████████▉ | 13079/14725 [6:38:47<51:18,  1.87s/it]
 89%|████████▉ | 13080/14725 [6:38:49<50:29,  1.84s/it]
                                                       

 89%|████████▉ | 13080/14725 [6:38:49<50:29,  1.84s/it]
 89%|████████▉ | 13081/14725 [6:38:51<50:07,  1.83s/it]
 89%|████████▉ | 13082/14725 [6:38:53<49:41,  1.81s/it]
 89%|████████▉ | 13083/14725 [6:38:54<49:08,  1.80s/it]
 89%|████████▉ | 13084/14725 [6:38:56<48:40,  1.78s/it]
 89%|████████▉ | 13085/14725 [6:38:58<48:52,  1.79s/it]
 89%|███�
+0: {'loss': 0.7936, 'grad_norm': 0.6606805123787062, 'learning_rate': 2e-05, 'epoch': 0.89}
+0: ��████▉ | 13086/14725 [6:39:00<49:03,  1.80s/it]
 89%|████████▉ | 13087/14725 [6:39:01<48:54,  1.79s/it]
 89%|████████▉ | 13088/14725 [6:39:03<49:41,  1.82s/it]
 89%|████████▉ | 13089/14725 [6:39:05<49:42,  1.82s/it]
 89%|████████▉ | 13090/14725 [6:39:07<49:20,  1.81s/it]
                                                       

 89%|████████▉ | 13090/14725 [6:39:07<49:20,  1.81s/it]
 89%|████████▉ | 13091/14725 [6:39:09<49:54,  1.83s/it]
 89%|████████▉ | 13092/14725 [6:39:11<49:45,  1.83s/it]
 89%|████████▉ | 13093/14725 [6:39:13<51:51,  1.91s/it]
 89%|████████▉ | 13094/14725 [6:39:15<51:25,  1.89s/it]
 89%|████████▉ | 13095/14725 [6:39:16<50:56,  1.88s/it]
 89%|████████▉ | 13096/14725 [6:39:18<50:24,  1.86s/it]
 89%|████████▉ | 13097/14725 [6:39:20<49:38,  1.83s/it]
 89%|█████
+0: {'loss': 0.7816, 'grad_norm': 0.6333447798296373, 'learning_rate': 2e-05, 'epoch': 0.89}
+0: ███▉ | 13098/14725 [6:39:22<50:06,  1.85s/it]
 89%|████████▉ | 13099/14725 [6:39:24<49:14,  1.82s/it]
 89%|████████▉ | 13100/14725 [6:39:25<49:03,  1.81s/it]
                                                       

 89%|████████▉ | 13100/14725 [6:39:25<49:03,  1.81s/it]
 89%|████████▉ | 13101/14725 [6:39:27<48:54,  1.81s/it]
 89%|████████▉ | 13102/14725 [6:39:29<48:27,  1.79s/it]
 89%|████████▉ | 13103/14725 [6:39:31<48:19,  1.79s/it]
 89%|████████▉ | 13104/14725 [6:39:33<48:14,  1.79s/it]
 89%|████████▉ | 13105/14725 [6:39:34<48:37,  1.80s/it]
 89%|████████▉ | 13106/14725 [6:39:36<48:25,  1.79s/it]
 89%|████████▉ | 13107/14725 [6:39:38<48:41,  1.81s/it]
 89%|████████▉ | 13108/14725 [6:39:40<48:52,  1.81s/it]
 89%|████████▉ | 13109/14725 [6:39:42<48:19,  1.79s/it]
 89%|██████�
+0: {'loss': 0.7848, 'grad_norm': 0.6711520198568679, 'learning_rate': 2e-05, 'epoch': 0.89}
+0: {'loss': 0.8031, 'grad_norm': 0.6826325578439099, 'learning_rate': 2e-05, 'epoch': 0.89}
+0: �█▉ | 13110/14725 [6:39:43<47:55,  1.78s/it]
                                                       

 89%|████████▉ | 13110/14725 [6:39:43<47:55,  1.78s/it]
 89%|████████▉ | 13111/14725 [6:39:45<47:43,  1.77s/it]
 89%|████████▉ | 13112/14725 [6:39:47<47:49,  1.78s/it]
 89%|████████▉ | 13113/14725 [6:39:49<48:13,  1.80s/it]
 89%|████████▉ | 13114/14725 [6:39:51<48:36,  1.81s/it]
 89%|████████▉ | 13115/14725 [6:39:52<48:24,  1.80s/it]
 89%|████████▉ | 13116/14725 [6:39:54<48:48,  1.82s/it]
 89%|████████▉ | 13117/14725 [6:39:56<48:24,  1.81s/it]
 89%|████████▉ | 13118/14725 [6:39:58<48:31,  1.81s/it]
 89%|████████▉ | 13119/14725 [6:40:00<48:57,  1.83s/it]
 89%|████████▉ | 13120/14725 [6:40:01<48:57,  1.83s/it]
                                                       

 89%|████████▉ | 13120/14725 
+0: {'loss': 0.7885, 'grad_norm': 0.6723935690847178, 'learning_rate': 2e-05, 'epoch': 0.89}
+0: [6:40:01<48:57,  1.83s/it]
 89%|████████▉ | 13121/14725 [6:40:03<49:11,  1.84s/it]
 89%|████████▉ | 13122/14725 [6:40:05<49:25,  1.85s/it]
 89%|████████▉ | 13123/14725 [6:40:07<49:09,  1.84s/it]
 89%|████████▉ | 13124/14725 [6:40:09<48:51,  1.83s/it]
 89%|████████▉ | 13125/14725 [6:40:11<48:18,  1.81s/it]
 89%|████████▉ | 13126/14725 [6:40:12<47:43,  1.79s/it]
 89%|████████▉ | 13127/14725 [6:40:14<47:59,  1.80s/it]
 89%|████████▉ | 13128/14725 [6:40:16<48:05,  1.81s/it]
 89%|████████▉ | 13129/14725 [6:40:18<48:41,  1.83s/it]
 89%|████████▉ | 13130/14725 [6:40:20<48:50,  1.84s/it]
                                                       

 89%|████████▉ | 13130/14725 [6:40:20<48:50,  1.84s/it]
 89%|████████▉ | 13131/14725 [6:40:22<48:50,  1.84s/it]
 89%|████████▉ | 13132/14725 [6:40
+0: {'loss': 0.8024, 'grad_norm': 0.6836829658574637, 'learning_rate': 2e-05, 'epoch': 0.89}
+0: :23<48:38,  1.83s/it]
 89%|████████▉ | 13133/14725 [6:40:25<48:38,  1.83s/it]
 89%|████████▉ | 13134/14725 [6:40:27<49:02,  1.85s/it]
 89%|████████▉ | 13135/14725 [6:40:29<48:25,  1.83s/it]
 89%|████████▉ | 13136/14725 [6:40:31<47:57,  1.81s/it]
 89%|████████▉ | 13137/14725 [6:40:32<48:00,  1.81s/it]
 89%|████████▉ | 13138/14725 [6:40:34<47:34,  1.80s/it]
 89%|████████▉ | 13139/14725 [6:40:36<47:39,  1.80s/it]
 89%|████████▉ | 13140/14725 [6:40:38<47:40,  1.80s/it]
                                                       

 89%|████████▉ | 13140/14725 [6:40:38<47:40,  1.80s/it]
 89%|████████▉ | 13141/14725 [6:40:40<47:57,  1.82s/it]
 89%|████████▉ | 13142/14725 [6:40:41<47:19,  1.79s/it]
 89%|████████▉ | 13143/14725 [6:40:43<48:02,  1.82s/it]
 89%|████████▉ | 13144/14725 [6:40:45<4
+0: {'loss': 0.7986, 'grad_norm': 0.6242396179414176, 'learning_rate': 2e-05, 'epoch': 0.89}
+0: 7:45,  1.81s/it]
 89%|████████▉ | 13145/14725 [6:40:47<48:12,  1.83s/it]
 89%|████████▉ | 13146/14725 [6:40:49<47:34,  1.81s/it]
 89%|████████▉ | 13147/14725 [6:40:51<47:31,  1.81s/it]
 89%|████████▉ | 13148/14725 [6:40:52<47:10,  1.79s/it]
 89%|████████▉ | 13149/14725 [6:40:54<47:20,  1.80s/it]
 89%|████████▉ | 13150/14725 [6:40:56<47:26,  1.81s/it]
                                                       

 89%|████████▉ | 13150/14725 [6:40:56<47:26,  1.81s/it]
 89%|████████▉ | 13151/14725 [6:40:58<47:43,  1.82s/it]
 89%|████████▉ | 13152/14725 [6:41:00<47:43,  1.82s/it]
 89%|████████▉ | 13153/14725 [6:41:01<47:15,  1.80s/it]
 89%|████████▉ | 13154/14725 [6:41:03<46:40,  1.78s/it]
 89%|████████▉ | 13155/14725 [6:41:05<46:39,  1.78s/it]
 89%|████████▉ | 13156/14725 [6:41:07<47:05,
+0: {'loss': 0.7778, 'grad_norm': 0.676867301114959, 'learning_rate': 2e-05, 'epoch': 0.89}
+0:   1.80s/it]
 89%|████████▉ | 13157/14725 [6:41:09<47:00,  1.80s/it]
 89%|████████▉ | 13158/14725 [6:41:10<47:11,  1.81s/it]
 89%|████████▉ | 13159/14725 [6:41:12<46:49,  1.79s/it]
 89%|████████▉ | 13160/14725 [6:41:14<46:40,  1.79s/it]
                                                       

 89%|████████▉ | 13160/14725 [6:41:14<46:40,  1.79s/it]
 89%|████████▉ | 13161/14725 [6:41:16<47:04,  1.81s/it]
 89%|████████▉ | 13162/14725 [6:41:18<46:34,  1.79s/it]
 89%|████████▉ | 13163/14725 [6:41:19<46:31,  1.79s/it]
 89%|████████▉ | 13164/14725 [6:41:21<46:56,  1.80s/it]
 89%|████████▉ | 13165/14725 [6:41:23<46:45,  1.80s/it]
 89%|████████▉ | 13166/14725 [6:41:25<46:20,  1.78s/it]
 89%|████████▉ | 13167/14725 [6:41:26<46:32,  1.79s/it]
 89%|████████▉ | 13168/14725 [6:41:28<46:31,  1.7
+0: {'loss': 0.7834, 'grad_norm': 0.6617386595639019, 'learning_rate': 2e-05, 'epoch': 0.89}
+0: 9s/it]
 89%|████████▉ | 13169/14725 [6:41:30<47:08,  1.82s/it]
 89%|████████▉ | 13170/14725 [6:41:32<47:03,  1.82s/it]
                                                       

 89%|████████▉ | 13170/14725 [6:41:32<47:03,  1.82s/it]
 89%|████████▉ | 13171/14725 [6:41:34<47:25,  1.83s/it]
 89%|████████▉ | 13172/14725 [6:41:36<48:08,  1.86s/it]
 89%|████████▉ | 13173/14725 [6:41:38<47:31,  1.84s/it]
 89%|████████▉ | 13174/14725 [6:41:39<47:13,  1.83s/it]
 89%|████████▉ | 13175/14725 [6:41:41<47:03,  1.82s/it]
 89%|████████▉ | 13176/14725 [6:41:43<47:06,  1.82s/it]
 89%|████████▉ | 13177/14725 [6:41:45<47:43,  1.85s/it]
 89%|████████▉ | 13178/14725 [6:41:47<46:58,  1.82s/it]
 90%|████████▉ | 13179/14725 [6:41:48<46:54,  1.82s/it]
 90%|████████▉ | 13180/14725 [6:41:50<46:41,  1.81s/it
+0: {'loss': 0.7845, 'grad_norm': 0.7137689694988048, 'learning_rate': 2e-05, 'epoch': 0.9}
+0: {'loss': 0.7766, 'grad_norm': 0.6389598978141158, 'learning_rate': 2e-05, 'epoch': 0.9}
+0: ]
                                                       

 90%|████████▉ | 13180/14725 [6:41:50<46:41,  1.81s/it]
 90%|████████▉ | 13181/14725 [6:41:52<46:37,  1.81s/it]
 90%|████████▉ | 13182/14725 [6:41:54<46:20,  1.80s/it]
 90%|████████▉ | 13183/14725 [6:41:56<46:27,  1.81s/it]
 90%|████████▉ | 13184/14725 [6:41:57<46:06,  1.80s/it]
 90%|████████▉ | 13185/14725 [6:41:59<46:03,  1.79s/it]
 90%|████████▉ | 13186/14725 [6:42:01<46:07,  1.80s/it]
 90%|████████▉ | 13187/14725 [6:42:03<45:43,  1.78s/it]
 90%|████████▉ | 13188/14725 [6:42:05<45:43,  1.78s/it]
 90%|████████▉ | 13189/14725 [6:42:06<46:10,  1.80s/it]
 90%|████████▉ | 13190/14725 [6:42:08<46:02,  1.80s/it]
                                                       

 90%|████████▉ | 13190/14725 [6:42:08<46:02,  1.80s/it]
 90%|█████
+0: {'loss': 0.7897, 'grad_norm': 0.7208660443356295, 'learning_rate': 2e-05, 'epoch': 0.9}
+0: ███▉ | 13191/14725 [6:42:10<45:57,  1.80s/it]
 90%|████████▉ | 13192/14725 [6:42:12<45:39,  1.79s/it]
 90%|████████▉ | 13193/14725 [6:42:14<45:45,  1.79s/it]
 90%|████████▉ | 13194/14725 [6:42:16<46:58,  1.84s/it]
 90%|████████▉ | 13195/14725 [6:42:17<46:25,  1.82s/it]
 90%|████████▉ | 13196/14725 [6:42:19<46:05,  1.81s/it]
 90%|████████▉ | 13197/14725 [6:42:21<46:12,  1.81s/it]
 90%|████████▉ | 13198/14725 [6:42:23<47:01,  1.85s/it]
 90%|████████▉ | 13199/14725 [6:42:25<46:29,  1.83s/it]
 90%|████████▉ | 13200/14725 [6:42:26<46:22,  1.82s/it]
                                                       

 90%|████████▉ | 13200/14725 [6:42:26<46:22,  1.82s/it]
 90%|████████▉ | 13201/14725 [6:42:28<46:23,  1.83s/it]
 90%|████████▉ | 13202/14725 [6:42:30<45:50,  1.81s/it]
 90%|██████�
+0: {'loss': 0.7935, 'grad_norm': 0.705639368905574, 'learning_rate': 2e-05, 'epoch': 0.9}
+0: �█▉ | 13203/14725 [6:42:32<45:35,  1.80s/it]
 90%|████████▉ | 13204/14725 [6:42:34<45:37,  1.80s/it]
 90%|████████▉ | 13205/14725 [6:42:35<45:59,  1.82s/it]
 90%|████████▉ | 13206/14725 [6:42:37<45:47,  1.81s/it]
 90%|████████▉ | 13207/14725 [6:42:39<45:47,  1.81s/it]
 90%|████████▉ | 13208/14725 [6:42:41<45:28,  1.80s/it]
 90%|████████▉ | 13209/14725 [6:42:43<45:03,  1.78s/it]
 90%|████████▉ | 13210/14725 [6:42:44<45:17,  1.79s/it]
                                                       

 90%|████████▉ | 13210/14725 [6:42:44<45:17,  1.79s/it]
 90%|████████▉ | 13211/14725 [6:42:46<44:57,  1.78s/it]
 90%|████████▉ | 13212/14725 [6:42:48<45:48,  1.82s/it]
 90%|████████▉ | 13213/14725 [6:42:50<47:38,  1.89s/it]
 90%|████████▉ | 13214/14725 [6:42:52<47:18,  1.88s/it]
 90%|████████�
+0: {'loss': 0.7929, 'grad_norm': 0.7014741758938348, 'learning_rate': 2e-05, 'epoch': 0.9}
+0: �� | 13215/14725 [6:42:54<46:34,  1.85s/it]
 90%|████████▉ | 13216/14725 [6:42:56<46:21,  1.84s/it]
 90%|████████▉ | 13217/14725 [6:42:57<45:46,  1.82s/it]
 90%|████████▉ | 13218/14725 [6:42:59<45:15,  1.80s/it]
 90%|████████▉ | 13219/14725 [6:43:01<45:22,  1.81s/it]
 90%|████████▉ | 13220/14725 [6:43:03<45:08,  1.80s/it]
                                                       

 90%|████████▉ | 13220/14725 [6:43:03<45:08,  1.80s/it]
 90%|████████▉ | 13221/14725 [6:43:04<44:56,  1.79s/it]
 90%|████████▉ | 13222/14725 [6:43:06<45:09,  1.80s/it]
 90%|████████▉ | 13223/14725 [6:43:08<44:56,  1.80s/it]
 90%|████████▉ | 13224/14725 [6:43:10<45:05,  1.80s/it]
 90%|████████▉ | 13225/14725 [6:43:12<45:31,  1.82s/it]
 90%|████████▉ | 13226/14725 [6:43:14<45:01,  1.80s/it]
 90%|████████▉ | 
+0: {'loss': 0.8038, 'grad_norm': 0.705553783574188, 'learning_rate': 2e-05, 'epoch': 0.9}
+0: 13227/14725 [6:43:15<44:45,  1.79s/it]
 90%|████████▉ | 13228/14725 [6:43:17<44:24,  1.78s/it]
 90%|████████▉ | 13229/14725 [6:43:19<44:27,  1.78s/it]
 90%|████████▉ | 13230/14725 [6:43:21<44:30,  1.79s/it]
                                                       

 90%|████████▉ | 13230/14725 [6:43:21<44:30,  1.79s/it]
 90%|████████▉ | 13231/14725 [6:43:22<44:34,  1.79s/it]
 90%|████████▉ | 13232/14725 [6:43:24<44:14,  1.78s/it]
 90%|████████▉ | 13233/14725 [6:43:26<44:11,  1.78s/it]
 90%|████████▉ | 13234/14725 [6:43:28<44:17,  1.78s/it]
 90%|████████▉ | 13235/14725 [6:43:30<44:33,  1.79s/it]
 90%|████████▉ | 13236/14725 [6:43:31<44:47,  1.80s/it]
 90%|████████▉ | 13237/14725 [6:43:33<44:33,  1.80s/it]
 90%|████████▉ | 13238/14725 [6:43:35<44:15,  1.79s/it]
 90%|████████▉ | 13239
+0: {'loss': 0.798, 'grad_norm': 0.6476137952985493, 'learning_rate': 2e-05, 'epoch': 0.9}
+0: /14725 [6:43:37<44:15,  1.79s/it]
 90%|████████▉ | 13240/14725 [6:43:39<44:44,  1.81s/it]
                                                       

 90%|████████▉ | 13240/14725 [6:43:39<44:44,  1.81s/it]
 90%|████████▉ | 13241/14725 [6:43:40<44:21,  1.79s/it]
 90%|████████▉ | 13242/14725 [6:43:42<44:27,  1.80s/it]
 90%|████████▉ | 13243/14725 [6:43:44<44:50,  1.82s/it]
 90%|████████▉ | 13244/14725 [6:43:46<44:52,  1.82s/it]
 90%|████████▉ | 13245/14725 [6:43:48<44:35,  1.81s/it]
 90%|████████▉ | 13246/14725 [6:43:49<44:22,  1.80s/it]
 90%|████████▉ | 13247/14725 [6:43:51<44:06,  1.79s/it]
 90%|████████▉ | 13248/14725 [6:43:53<44:53,  1.82s/it]
 90%|████████▉ | 13249/14725 [6:43:55<44:42,  1.82s/it]
 90%|████████▉ | 13250/14725 [6:43:57<45:01,  1.83s/it]
                                             
+0: {'loss': 0.7892, 'grad_norm': 0.661812177343065, 'learning_rate': 2e-05, 'epoch': 0.9}
+0: {'loss': 0.8009, 'grad_norm': 0.6648765017537756, 'learning_rate': 2e-05, 'epoch': 0.9}
+0:           

 90%|████████▉ | 13250/14725 [6:43:57<45:01,  1.83s/it]
 90%|████████▉ | 13251/14725 [6:43:59<44:48,  1.82s/it]
 90%|████████▉ | 13252/14725 [6:44:00<44:38,  1.82s/it]
 90%|█████████ | 13253/14725 [6:44:02<44:31,  1.81s/it]
 90%|█████████ | 13254/14725 [6:44:04<44:02,  1.80s/it]
 90%|█████████ | 13255/14725 [6:44:06<44:13,  1.81s/it]
 90%|█████████ | 13256/14725 [6:44:08<44:31,  1.82s/it]
 90%|█████████ | 13257/14725 [6:44:10<45:11,  1.85s/it]
 90%|█████████ | 13258/14725 [6:44:11<44:34,  1.82s/it]
 90%|█████████ | 13259/14725 [6:44:13<44:18,  1.81s/it]
 90%|█████████ | 13260/14725 [6:44:15<44:47,  1.83s/it]
                                                       

 90%|█████████ | 13260/14725 [6:44:15<44:47,  1.83s/it]
 90%|█████████ | 13261/14725 [6:44:17<44:48,  1.8
+0: {'loss': 0.8011, 'grad_norm': 0.6856513746834052, 'learning_rate': 2e-05, 'epoch': 0.9}
+0: 4s/it]
 90%|█████████ | 13262/14725 [6:44:19<44:48,  1.84s/it]
 90%|█████████ | 13263/14725 [6:44:20<44:02,  1.81s/it]
 90%|█████████ | 13264/14725 [6:44:22<43:34,  1.79s/it]
 90%|█████████ | 13265/14725 [6:44:24<43:22,  1.78s/it]
 90%|█████████ | 13266/14725 [6:44:26<43:19,  1.78s/it]
 90%|█████████ | 13267/14725 [6:44:28<43:56,  1.81s/it]
 90%|█████████ | 13268/14725 [6:44:29<44:37,  1.84s/it]
 90%|█████████ | 13269/14725 [6:44:31<44:18,  1.83s/it]
 90%|█████████ | 13270/14725 [6:44:33<44:14,  1.82s/it]
                                                       

 90%|█████████ | 13270/14725 [6:44:33<44:14,  1.82s/it]
 90%|█████████ | 13271/14725 [6:44:35<43:45,  1.81s/it]
 90%|█████████ | 13272/14725 [6:44:37<43:57,  1.82s/it]
 90%|█████████ | 13273/14725 [6:44:38<44:03,  1.82s/it
+0: {'loss': 0.801, 'grad_norm': 0.7180271399299583, 'learning_rate': 2e-05, 'epoch': 0.9}
+0: ]
 90%|█████████ | 13274/14725 [6:44:40<43:57,  1.82s/it]
 90%|█████████ | 13275/14725 [6:44:42<43:46,  1.81s/it]
 90%|█████████ | 13276/14725 [6:44:44<43:15,  1.79s/it]
 90%|█████████ | 13277/14725 [6:44:46<43:16,  1.79s/it]
 90%|█████████ | 13278/14725 [6:44:47<43:32,  1.81s/it]
 90%|█████████ | 13279/14725 [6:44:49<43:36,  1.81s/it]
 90%|█████████ | 13280/14725 [6:44:51<43:22,  1.80s/it]
                                                       

 90%|█████████ | 13280/14725 [6:44:51<43:22,  1.80s/it]
 90%|█████████ | 13281/14725 [6:44:53<43:14,  1.80s/it]
 90%|█████████ | 13282/14725 [6:44:55<42:52,  1.78s/it]
 90%|█████████ | 13283/14725 [6:44:56<42:49,  1.78s/it]
 90%|█████████ | 13284/14725 [6:44:58<42:48,  1.78s/it]
 90%|█████████ | 13285/14725 [6:45:00<42:34,  1.77s/it]
 90
+0: {'loss': 0.7856, 'grad_norm': 0.6791123611048007, 'learning_rate': 2e-05, 'epoch': 0.9}
+0: %|█████████ | 13286/14725 [6:45:02<42:49,  1.79s/it]
 90%|█████████ | 13287/14725 [6:45:04<43:25,  1.81s/it]
 90%|█████████ | 13288/14725 [6:45:05<43:26,  1.81s/it]
 90%|█████████ | 13289/14725 [6:45:07<43:40,  1.82s/it]
 90%|█████████ | 13290/14725 [6:45:09<43:15,  1.81s/it]
                                                       

 90%|█████████ | 13290/14725 [6:45:09<43:15,  1.81s/it]
 90%|█████████ | 13291/14725 [6:45:11<43:10,  1.81s/it]
 90%|█████████ | 13292/14725 [6:45:13<42:56,  1.80s/it]
 90%|█████████ | 13293/14725 [6:45:14<42:47,  1.79s/it]
 90%|█████████ | 13294/14725 [6:45:16<42:46,  1.79s/it]
 90%|█████████ | 13295/14725 [6:45:18<42:47,  1.80s/it]
 90%|█████████ | 13296/14725 [6:45:20<42:50,  1.80s/it]
 90%|█████████ | 13297/14725 [6:45:22<42:58,  1.81s/it]
 90%|█
+0: {'loss': 0.7767, 'grad_norm': 0.6517959662977234, 'learning_rate': 2e-05, 'epoch': 0.9}
+0: ████████ | 13298/14725 [6:45:23<42:55,  1.80s/it]
 90%|█████████ | 13299/14725 [6:45:25<42:52,  1.80s/it]
 90%|█████████ | 13300/14725 [6:45:27<42:31,  1.79s/it]
                                                       

 90%|█████████ | 13300/14725 [6:45:27<42:31,  1.79s/it]
 90%|█████████ | 13301/14725 [6:45:29<43:14,  1.82s/it]
 90%|█████████ | 13302/14725 [6:45:31<43:39,  1.84s/it]
 90%|█████████ | 13303/14725 [6:45:33<42:59,  1.81s/it]
 90%|█████████ | 13304/14725 [6:45:34<42:45,  1.81s/it]
 90%|█████████ | 13305/14725 [6:45:36<42:51,  1.81s/it]
 90%|█████████ | 13306/14725 [6:45:38<43:03,  1.82s/it]
 90%|█████████ | 13307/14725 [6:45:40<42:57,  1.82s/it]
 90%|█████████ | 13308/14725 [6:45:42<42:46,  1.81s/it]
 90%|█████████ | 13309/14725 [6:45:43<43:17,  1.83s/it]
 90%|██�
+0: {'loss': 0.8172, 'grad_norm': 0.7269630972427767, 'learning_rate': 2e-05, 'epoch': 0.9}
+0: {'loss': 0.8098, 'grad_norm': 0.6469054265907137, 'learning_rate': 2e-05, 'epoch': 0.9}
+0: �██████ | 13310/14725 [6:45:45<42:54,  1.82s/it]
                                                       

 90%|█████████ | 13310/14725 [6:45:45<42:54,  1.82s/it]
 90%|█████████ | 13311/14725 [6:45:47<43:06,  1.83s/it]
 90%|█████████ | 13312/14725 [6:45:49<43:19,  1.84s/it]
 90%|█████████ | 13313/14725 [6:45:51<43:18,  1.84s/it]
 90%|█████████ | 13314/14725 [6:45:53<42:58,  1.83s/it]
 90%|█████████ | 13315/14725 [6:45:54<43:10,  1.84s/it]
 90%|█████████ | 13316/14725 [6:45:56<43:43,  1.86s/it]
 90%|█████████ | 13317/14725 [6:45:58<43:07,  1.84s/it]
 90%|█████████ | 13318/14725 [6:46:00<42:42,  1.82s/it]
 90%|█████████ | 13319/14725 [6:46:02<42:44,  1.82s/it]
 90%|█████████ | 13320/14725 [6:46:04<42:39,  1.82s/it]
                                                       

 90%|█████████ | 
+0: {'loss': 0.806, 'grad_norm': 0.6360286462064625, 'learning_rate': 2e-05, 'epoch': 0.91}
+0: 13320/14725 [6:46:04<42:39,  1.82s/it]
 90%|█████████ | 13321/14725 [6:46:05<42:20,  1.81s/it]
 90%|█████████ | 13322/14725 [6:46:07<42:15,  1.81s/it]
 90%|█████████ | 13323/14725 [6:46:09<41:55,  1.79s/it]
 90%|█████████ | 13324/14725 [6:46:11<42:03,  1.80s/it]
 90%|█████████ | 13325/14725 [6:46:13<42:17,  1.81s/it]
 90%|█████████ | 13326/14725 [6:46:14<42:15,  1.81s/it]
 91%|█████████ | 13327/14725 [6:46:16<42:09,  1.81s/it]
 91%|█████████ | 13328/14725 [6:46:18<42:11,  1.81s/it]
 91%|█████████ | 13329/14725 [6:46:20<42:03,  1.81s/it]
 91%|█████████ | 13330/14725 [6:46:22<42:02,  1.81s/it]
                                                       

 91%|█████████ | 13330/14725 [6:46:22<42:02,  1.81s/it]
 91%|█████████ | 13331/14725 [6:46:24<42:17,  1.82s/it]
 91%|█████████ | 13332
+0: {'loss': 0.7999, 'grad_norm': 0.6374961110125642, 'learning_rate': 2e-05, 'epoch': 0.91}
+0: /14725 [6:46:25<42:13,  1.82s/it]
 91%|█████████ | 13333/14725 [6:46:27<42:23,  1.83s/it]
 91%|█████████ | 13334/14725 [6:46:29<42:23,  1.83s/it]
 91%|█████████ | 13335/14725 [6:46:31<41:54,  1.81s/it]
 91%|█████████ | 13336/14725 [6:46:33<41:45,  1.80s/it]
 91%|█████████ | 13337/14725 [6:46:34<42:13,  1.82s/it]
 91%|█████████ | 13338/14725 [6:46:36<41:44,  1.81s/it]
 91%|█████████ | 13339/14725 [6:46:38<42:29,  1.84s/it]
 91%|█████████ | 13340/14725 [6:46:40<42:10,  1.83s/it]
                                                       

 91%|█████████ | 13340/14725 [6:46:40<42:10,  1.83s/it]
 91%|█████████ | 13341/14725 [6:46:42<42:13,  1.83s/it]
 91%|█████████ | 13342/14725 [6:46:44<41:46,  1.81s/it]
 91%|█████████ | 13343/14725 [6:46:45<41:53,  1.82s/it]
 91%|█████████ | 13344/1472
+0: {'loss': 0.7809, 'grad_norm': 0.6500054747221803, 'learning_rate': 2e-05, 'epoch': 0.91}
+0: 5 [6:46:47<41:54,  1.82s/it]
 91%|█████████ | 13345/14725 [6:46:49<41:50,  1.82s/it]
 91%|█████████ | 13346/14725 [6:46:51<41:46,  1.82s/it]
 91%|█████████ | 13347/14725 [6:46:53<41:25,  1.80s/it]
 91%|█████████ | 13348/14725 [6:46:54<41:09,  1.79s/it]
 91%|█████████ | 13349/14725 [6:46:56<41:42,  1.82s/it]
 91%|█████████ | 13350/14725 [6:46:58<41:51,  1.83s/it]
                                                       

 91%|█████████ | 13350/14725 [6:46:58<41:51,  1.83s/it]
 91%|█████████ | 13351/14725 [6:47:00<42:30,  1.86s/it]
 91%|█████████ | 13352/14725 [6:47:02<41:58,  1.83s/it]
 91%|█████████ | 13353/14725 [6:47:04<41:22,  1.81s/it]
 91%|█████████ | 13354/14725 [6:47:05<40:56,  1.79s/it]
 91%|█████████ | 13355/14725 [6:47:07<40:54,  1.79s/it]
 91%|█████████ | 13356/14725 [6:
+0: {'loss': 0.7857, 'grad_norm': 0.6466591369062692, 'learning_rate': 2e-05, 'epoch': 0.91}
+0: 47:09<40:57,  1.79s/it]
 91%|█████████ | 13357/14725 [6:47:11<41:18,  1.81s/it]
 91%|█████████ | 13358/14725 [6:47:13<41:22,  1.82s/it]
 91%|█████████ | 13359/14725 [6:47:14<41:18,  1.81s/it]
 91%|█████████ | 13360/14725 [6:47:16<41:18,  1.82s/it]
                                                       

 91%|█████████ | 13360/14725 [6:47:16<41:18,  1.82s/it]
 91%|█████████ | 13361/14725 [6:47:18<41:12,  1.81s/it]
 91%|█████████ | 13362/14725 [6:47:20<41:46,  1.84s/it]
 91%|█████████ | 13363/14725 [6:47:22<41:40,  1.84s/it]
 91%|█████████ | 13364/14725 [6:47:24<41:36,  1.83s/it]
 91%|█████████ | 13365/14725 [6:47:25<41:43,  1.84s/it]
 91%|█████████ | 13366/14725 [6:47:27<41:06,  1.81s/it]
 91%|█████████ | 13367/14725 [6:47:29<40:53,  1.81s/it]
 91%|█████████ | 13368/14725 [6:47:31
+0: {'loss': 0.7806, 'grad_norm': 0.6581397211436748, 'learning_rate': 2e-05, 'epoch': 0.91}
+0: <40:44,  1.80s/it]
 91%|█████████ | 13369/14725 [6:47:33<40:44,  1.80s/it]
 91%|█████████ | 13370/14725 [6:47:34<40:20,  1.79s/it]
                                                       

 91%|█████████ | 13370/14725 [6:47:34<40:20,  1.79s/it]
 91%|█████████ | 13371/14725 [6:47:36<40:02,  1.77s/it]
 91%|█████████ | 13372/14725 [6:47:38<40:24,  1.79s/it]
 91%|█████████ | 13373/14725 [6:47:40<41:14,  1.83s/it]
 91%|█████████ | 13374/14725 [6:47:42<40:48,  1.81s/it]
 91%|█████████ | 13375/14725 [6:47:43<40:43,  1.81s/it]
 91%|█████████ | 13376/14725 [6:47:45<40:58,  1.82s/it]
 91%|█████████ | 13377/14725 [6:47:47<40:47,  1.82s/it]
 91%|█████████ | 13378/14725 [6:47:49<40:22,  1.80s/it]
 91%|█████████ | 13379/14725 [6:47:51<40:06,  1.79s/it]
 91%|█████████ | 13380/14725 [6:47:52<39:5
+0: {'loss': 0.7904, 'grad_norm': 0.6785574987282281, 'learning_rate': 2e-05, 'epoch': 0.91}
+0: {'loss': 0.796, 'grad_norm': 0.6772130699662549, 'learning_rate': 2e-05, 'epoch': 0.91}
+0: 4,  1.78s/it]
                                                       

 91%|█████████ | 13380/14725 [6:47:52<39:54,  1.78s/it]
 91%|█████████ | 13381/14725 [6:47:54<39:44,  1.77s/it]
 91%|█████████ | 13382/14725 [6:47:56<39:45,  1.78s/it]
 91%|█████████ | 13383/14725 [6:47:58<40:07,  1.79s/it]
 91%|█████████ | 13384/14725 [6:47:59<39:49,  1.78s/it]
 91%|█████████ | 13385/14725 [6:48:01<39:43,  1.78s/it]
 91%|█████████ | 13386/14725 [6:48:03<39:43,  1.78s/it]
 91%|█████████ | 13387/14725 [6:48:05<40:10,  1.80s/it]
 91%|█████████ | 13388/14725 [6:48:07<40:04,  1.80s/it]
 91%|█████████ | 13389/14725 [6:48:08<40:33,  1.82s/it]
 91%|█████████ | 13390/14725 [6:48:10<40:21,  1.81s/it]
                                                       

 91%|█████████ | 13390/14725 [6:48:10<40:21,  1.81s/it]
 91%|█
+0: {'loss': 0.7912, 'grad_norm': 0.6717004179624414, 'learning_rate': 2e-05, 'epoch': 0.91}
+0: ████████ | 13391/14725 [6:48:12<40:10,  1.81s/it]
 91%|█████████ | 13392/14725 [6:48:14<39:54,  1.80s/it]
 91%|█████████ | 13393/14725 [6:48:16<39:36,  1.78s/it]
 91%|█████████ | 13394/14725 [6:48:17<40:08,  1.81s/it]
 91%|█████████ | 13395/14725 [6:48:19<40:18,  1.82s/it]
 91%|█████████ | 13396/14725 [6:48:21<39:45,  1.80s/it]
 91%|█████████ | 13397/14725 [6:48:23<39:37,  1.79s/it]
 91%|█████████ | 13398/14725 [6:48:25<39:31,  1.79s/it]
 91%|█████████ | 13399/14725 [6:48:26<39:12,  1.77s/it]
 91%|█████████ | 13400/14725 [6:48:28<40:27,  1.83s/it]
                                                       

 91%|█████████ | 13400/14725 [6:48:28<40:27,  1.83s/it]
 91%|█████████ | 13401/14725 [6:48:30<39:56,  1.81s/it]
 91%|█████████ | 13402/14725 [6:48:32<40:14,  1.83s/it]
 91%|██�
+0: {'loss': 0.7807, 'grad_norm': 0.6559922246868682, 'learning_rate': 2e-05, 'epoch': 0.91}
+0: �██████ | 13403/14725 [6:48:34<40:46,  1.85s/it]
 91%|█████████ | 13404/14725 [6:48:36<40:20,  1.83s/it]
 91%|█████████ | 13405/14725 [6:48:37<40:26,  1.84s/it]
 91%|█████████ | 13406/14725 [6:48:39<39:58,  1.82s/it]
 91%|█████████ | 13407/14725 [6:48:41<40:25,  1.84s/it]
 91%|█████████ | 13408/14725 [6:48:43<39:42,  1.81s/it]
 91%|█████████ | 13409/14725 [6:48:45<39:39,  1.81s/it]
 91%|█████████ | 13410/14725 [6:48:47<39:42,  1.81s/it]
                                                       

 91%|█████████ | 13410/14725 [6:48:47<39:42,  1.81s/it]
 91%|█████████ | 13411/14725 [6:48:48<39:17,  1.79s/it]
 91%|█████████ | 13412/14725 [6:48:50<39:05,  1.79s/it]
 91%|█████████ | 13413/14725 [6:48:52<39:29,  1.81s/it]
 91%|█████████ | 13414/14725 [6:48:54<39:50,  1.82s/it]
 91%|████�
+0: {'loss': 0.7984, 'grad_norm': 0.6697645995723448, 'learning_rate': 2e-05, 'epoch': 0.91}
+0: ��████ | 13415/14725 [6:48:56<39:34,  1.81s/it]
 91%|█████████ | 13416/14725 [6:48:57<39:41,  1.82s/it]
 91%|█████████ | 13417/14725 [6:48:59<39:20,  1.80s/it]
 91%|█████████ | 13418/14725 [6:49:01<39:36,  1.82s/it]
 91%|█████████ | 13419/14725 [6:49:03<39:30,  1.81s/it]
 91%|█████████ | 13420/14725 [6:49:05<39:25,  1.81s/it]
                                                       

 91%|█████████ | 13420/14725 [6:49:05<39:25,  1.81s/it]
 91%|█████████ | 13421/14725 [6:49:06<39:29,  1.82s/it]
 91%|█████████ | 13422/14725 [6:49:08<39:40,  1.83s/it]
 91%|█████████ | 13423/14725 [6:49:10<39:40,  1.83s/it]
 91%|█████████ | 13424/14725 [6:49:12<39:11,  1.81s/it]
 91%|█████████ | 13425/14725 [6:49:14<39:13,  1.81s/it]
 91%|█████████ | 13426/14725 [6:49:15<38:47,  1.79s/it]
 91%|██████
+0: {'loss': 0.8026, 'grad_norm': 0.6864937434563467, 'learning_rate': 2e-05, 'epoch': 0.91}
+0: ███ | 13427/14725 [6:49:17<38:54,  1.80s/it]
 91%|█████████ | 13428/14725 [6:49:19<38:41,  1.79s/it]
 91%|█████████ | 13429/14725 [6:49:21<38:36,  1.79s/it]
 91%|█████████ | 13430/14725 [6:49:23<39:05,  1.81s/it]
                                                       

 91%|█████████ | 13430/14725 [6:49:23<39:05,  1.81s/it]
 91%|█████████ | 13431/14725 [6:49:24<38:42,  1.80s/it]
 91%|█████████ | 13432/14725 [6:49:26<38:37,  1.79s/it]
 91%|█████████ | 13433/14725 [6:49:28<38:55,  1.81s/it]
 91%|█████████ | 13434/14725 [6:49:30<38:33,  1.79s/it]
 91%|█████████ | 13435/14725 [6:49:32<38:23,  1.79s/it]
 91%|█████████ | 13436/14725 [6:49:33<38:16,  1.78s/it]
 91%|█████████▏| 13437/14725 [6:49:35<38:51,  1.81s/it]
 91%|█████████▏| 13438/14725 [6:49:37<38:50,  1.81s/it]
 91%|██████�
+0: {'loss': 0.7853, 'grad_norm': 0.6706361750438163, 'learning_rate': 2e-05, 'epoch': 0.91}
+0: ��██▏| 13439/14725 [6:49:39<38:58,  1.82s/it]
 91%|█████████▏| 13440/14725 [6:49:41<39:02,  1.82s/it]
                                                       

 91%|█████████▏| 13440/14725 [6:49:41<39:02,  1.82s/it]
 91%|█████████▏| 13441/14725 [6:49:43<38:55,  1.82s/it]
 91%|█████████▏| 13442/14725 [6:49:44<39:09,  1.83s/it]
 91%|█████████▏| 13443/14725 [6:49:46<38:52,  1.82s/it]
 91%|█████████▏| 13444/14725 [6:49:48<39:06,  1.83s/it]
 91%|█████████▏| 13445/14725 [6:49:50<38:41,  1.81s/it]
 91%|█████████▏| 13446/14725 [6:49:52<38:27,  1.80s/it]
 91%|█████████▏| 13447/14725 [6:49:53<38:17,  1.80s/it]
 91%|█████████▏| 13448/14725 [6:49:55<38:07,  1.79s/it]
 91%|█████████▏| 13449/14725 [6:49:57<38:41,  1.82s/it]
 91%|█████████▏| 13450/14725 [6:49:59<38:10,  1.80s/it]
   
+0: {'loss': 0.7694, 'grad_norm': 0.6940813494779254, 'learning_rate': 2e-05, 'epoch': 0.91}
+0: {'loss': 0.7893, 'grad_norm': 0.667067031651937, 'learning_rate': 2e-05, 'epoch': 0.91}
+0:                                                     

 91%|█████████▏| 13450/14725 [6:49:59<38:10,  1.80s/it]
 91%|█████████▏| 13451/14725 [6:50:01<38:03,  1.79s/it]
 91%|█████████▏| 13452/14725 [6:50:02<38:31,  1.82s/it]
 91%|█████████▏| 13453/14725 [6:50:04<38:19,  1.81s/it]
 91%|█████████▏| 13454/14725 [6:50:06<38:16,  1.81s/it]
 91%|█████████▏| 13455/14725 [6:50:08<38:14,  1.81s/it]
 91%|█████████▏| 13456/14725 [6:50:10<38:21,  1.81s/it]
 91%|█████████▏| 13457/14725 [6:50:11<37:52,  1.79s/it]
 91%|█████████▏| 13458/14725 [6:50:13<37:45,  1.79s/it]
 91%|█████████▏| 13459/14725 [6:50:15<37:42,  1.79s/it]
 91%|█████████▏| 13460/14725 [6:50:17<37:53,  1.80s/it]
                                                       

 91%|█████████▏| 13460/14725 [6:50:17<37:53,  1.80s/it]
 
+0: {'loss': 0.7905, 'grad_norm': 0.688841681017196, 'learning_rate': 2e-05, 'epoch': 0.91}
+0: 91%|█████████▏| 13461/14725 [6:50:19<37:41,  1.79s/it]
 91%|█████████▏| 13462/14725 [6:50:20<38:14,  1.82s/it]
 91%|█████████▏| 13463/14725 [6:50:22<38:12,  1.82s/it]
 91%|█████████▏| 13464/14725 [6:50:24<37:53,  1.80s/it]
 91%|█████████▏| 13465/14725 [6:50:26<38:01,  1.81s/it]
 91%|█████████▏| 13466/14725 [6:50:28<37:59,  1.81s/it]
 91%|█████████▏| 13467/14725 [6:50:30<38:03,  1.82s/it]
 91%|█████████▏| 13468/14725 [6:50:31<38:27,  1.84s/it]
 91%|█████████▏| 13469/14725 [6:50:33<38:09,  1.82s/it]
 91%|█████████▏| 13470/14725 [6:50:35<37:55,  1.81s/it]
                                                       

 91%|█████████▏| 13470/14725 [6:50:35<37:55,  1.81s/it]
 91%|█████████▏| 13471/14725 [6:50:37<37:58,  1.82s/it]
 91%|█████████▏| 13472/14725 [6:50:3
+0: {'loss': 0.787, 'grad_norm': 0.6589877897806132, 'learning_rate': 2e-05, 'epoch': 0.92}
+0: 9<37:40,  1.80s/it]
 91%|█████████▏| 13473/14725 [6:50:40<37:31,  1.80s/it]
 92%|█████████▏| 13474/14725 [6:50:42<38:11,  1.83s/it]
 92%|█████████▏| 13475/14725 [6:50:44<37:58,  1.82s/it]
 92%|█████████▏| 13476/14725 [6:50:46<38:18,  1.84s/it]
 92%|█████████▏| 13477/14725 [6:50:48<38:09,  1.83s/it]
 92%|█████████▏| 13478/14725 [6:50:50<37:48,  1.82s/it]
 92%|█████████▏| 13479/14725 [6:50:51<37:35,  1.81s/it]
 92%|█████████▏| 13480/14725 [6:50:53<37:19,  1.80s/it]
                                                       

 92%|█████████▏| 13480/14725 [6:50:53<37:19,  1.80s/it]
 92%|█████████▏| 13481/14725 [6:50:55<37:54,  1.83s/it]
 92%|█████████▏| 13482/14725 [6:50:57<37:29,  1.81s/it]
 92%|█████████▏| 13483/14725 [6:50:59<37:26,  1.81s/it]
 92%|█████████▏
+0: {'loss': 0.7881, 'grad_norm': 0.6858049778652681, 'learning_rate': 2e-05, 'epoch': 0.92}
+0: | 13484/14725 [6:51:00<37:04,  1.79s/it]
 92%|█████████▏| 13485/14725 [6:51:02<37:29,  1.81s/it]
 92%|█████████▏| 13486/14725 [6:51:04<37:40,  1.82s/it]
 92%|█████████▏| 13487/14725 [6:51:06<37:11,  1.80s/it]
 92%|█████████▏| 13488/14725 [6:51:08<36:59,  1.79s/it]
 92%|█████████▏| 13489/14725 [6:51:09<36:54,  1.79s/it]
 92%|█████████▏| 13490/14725 [6:51:11<36:57,  1.80s/it]
                                                       

 92%|█████████▏| 13490/14725 [6:51:11<36:57,  1.80s/it]
 92%|█████████▏| 13491/14725 [6:51:13<37:04,  1.80s/it]
 92%|█████████▏| 13492/14725 [6:51:15<37:02,  1.80s/it]
 92%|█████████▏| 13493/14725 [6:51:17<37:01,  1.80s/it]
 92%|█████████▏| 13494/14725 [6:51:18<36:45,  1.79s/it]
 92%|█████████▏| 13495/14725 [6:51:20<36:58,  1.80s/it]
 92%|███
+0: {'loss': 0.7775, 'grad_norm': 0.6471244870118812, 'learning_rate': 2e-05, 'epoch': 0.92}
+0: ██████▏| 13496/14725 [6:51:22<37:32,  1.83s/it]
 92%|█████████▏| 13497/14725 [6:51:24<37:04,  1.81s/it]
 92%|█████████▏| 13498/14725 [6:51:26<36:48,  1.80s/it]
 92%|█████████▏| 13499/14725 [6:51:27<36:37,  1.79s/it]
 92%|█████████▏| 13500/14725 [6:51:29<36:37,  1.79s/it]
                                                       

 92%|█████████▏| 13500/14725 [6:51:29<36:37,  1.79s/it]
 92%|█████████▏| 13501/14725 [6:51:31<36:34,  1.79s/it]
 92%|█████████▏| 13502/14725 [6:51:33<36:41,  1.80s/it]
 92%|█████████▏| 13503/14725 [6:51:35<37:02,  1.82s/it]
 92%|█████████▏| 13504/14725 [6:51:36<36:45,  1.81s/it]
 92%|█████████▏| 13505/14725 [6:51:38<36:42,  1.80s/it]
 92%|█████████▏| 13506/14725 [6:51:40<36:25,  1.79s/it]
 92%|█████████▏| 13507/14725 [6:51:42<36:43,  1.8
+0: {'loss': 0.7882, 'grad_norm': 0.6182455010892621, 'learning_rate': 2e-05, 'epoch': 0.92}
+0: 1s/it]
 92%|█████████▏| 13508/14725 [6:51:44<36:31,  1.80s/it]
 92%|█████████▏| 13509/14725 [6:51:45<36:33,  1.80s/it]
 92%|█████████▏| 13510/14725 [6:51:47<36:20,  1.79s/it]
                                                       

 92%|█████████▏| 13510/14725 [6:51:47<36:20,  1.79s/it]
 92%|█████████▏| 13511/14725 [6:51:49<36:31,  1.81s/it]
 92%|█████████▏| 13512/14725 [6:51:51<36:21,  1.80s/it]
 92%|█████████▏| 13513/14725 [6:51:53<36:28,  1.81s/it]
 92%|█████████▏| 13514/14725 [6:51:54<36:08,  1.79s/it]
 92%|█████████▏| 13515/14725 [6:51:56<36:23,  1.80s/it]
 92%|█████████▏| 13516/14725 [6:51:58<36:24,  1.81s/it]
 92%|█████████▏| 13517/14725 [6:52:00<36:29,  1.81s/it]
 92%|█████████▏| 13518/14725 [6:52:02<36:25,  1.81s/it]
 92%|█████████▏| 13519/14725
+0: {'loss': 0.7789, 'grad_norm': 0.6283104662285494, 'learning_rate': 2e-05, 'epoch': 0.92}
+0:  [6:52:03<36:16,  1.80s/it]
 92%|█████████▏| 13520/14725 [6:52:05<35:49,  1.78s/it]
                                                       

 92%|█████████▏| 13520/14725 [6:52:05<35:49,  1.78s/it]
 92%|█████████▏| 13521/14725 [6:52:07<35:39,  1.78s/it]
 92%|█████████▏| 13522/14725 [6:52:09<35:56,  1.79s/it]
 92%|█████████▏| 13523/14725 [6:52:11<35:55,  1.79s/it]
 92%|█████████▏| 13524/14725 [6:52:12<35:56,  1.80s/it]
 92%|█████████▏| 13525/14725 [6:52:14<35:50,  1.79s/it]
 92%|█████████▏| 13526/14725 [6:52:16<36:10,  1.81s/it]
 92%|███████��█▏| 13527/14725 [6:52:18<36:08,  1.81s/it]
 92%|█████████▏| 13528/14725 [6:52:20<35:47,  1.79s/it]
 92%|█████████▏| 13529/14725 [6:52:21<35:39,  1.79s/it]
 92%|█████████▏| 13530/14725 [6:52:23<36:08,  1.81s/it]
                           
+0: {'loss': 0.7881, 'grad_norm': 0.6843511684680406, 'learning_rate': 2e-05, 'epoch': 0.92}
+0: {'loss': 0.7936, 'grad_norm': 0.6525073831063636, 'learning_rate': 2e-05, 'epoch': 0.92}
+0:                             

 92%|█████████▏| 13530/14725 [6:52:23<36:08,  1.81s/it]
 92%|█████████▏| 13531/14725 [6:52:25<36:25,  1.83s/it]
 92%|█████████▏| 13532/14725 [6:52:27<36:19,  1.83s/it]
 92%|█████████▏| 13533/14725 [6:52:29<36:11,  1.82s/it]
 92%|█████████▏| 13534/14725 [6:52:31<35:57,  1.81s/it]
 92%|█████████▏| 13535/14725 [6:52:32<35:44,  1.80s/it]
 92%|█████████▏| 13536/14725 [6:52:34<35:57,  1.81s/it]
 92%|█████████▏| 13537/14725 [6:52:36<35:38,  1.80s/it]
 92%|█████████▏| 13538/14725 [6:52:38<36:44,  1.86s/it]
 92%|█████████▏| 13539/14725 [6:52:40<36:13,  1.83s/it]
 92%|█████████▏| 13540/14725 [6:52:42<36:00,  1.82s/it]
                                                       

 92%|█████████▏| 13540/14725 [6:52:42<36:00,  1.82s/it]
 92%|██████�
+0: {'loss': 0.7845, 'grad_norm': 0.6833705197490315, 'learning_rate': 2e-05, 'epoch': 0.92}
+0: �██▏| 13541/14725 [6:52:43<36:12,  1.83s/it]
 92%|█████████▏| 13542/14725 [6:52:45<35:57,  1.82s/it]
 92%|█████████▏| 13543/14725 [6:52:47<36:00,  1.83s/it]
 92%|█████████▏| 13544/14725 [6:52:49<35:57,  1.83s/it]
 92%|█████████▏| 13545/14725 [6:52:51<35:38,  1.81s/it]
 92%|█████████▏| 13546/14725 [6:52:52<35:33,  1.81s/it]
 92%|█████████▏| 13547/14725 [6:52:54<36:21,  1.85s/it]
 92%|█████████▏| 13548/14725 [6:52:56<36:43,  1.87s/it]
 92%|█████████▏| 13549/14725 [6:52:58<36:19,  1.85s/it]
 92%|█████████▏| 13550/14725 [6:53:00<35:45,  1.83s/it]
                                                       

 92%|█████████▏| 13550/14725 [6:53:00<35:45,  1.83s/it]
 92%|█████████▏| 13551/14725 [6:53:02<36:00,  1.84s/it]
 92%|█████████▏| 13552/14725 [6:53:04<35:50,  1.83s/it]
 92%
+0: {'loss': 0.7993, 'grad_norm': 0.6683893955416129, 'learning_rate': 2e-05, 'epoch': 0.92}
+0: |█████████▏| 13553/14725 [6:53:05<35:48,  1.83s/it]
 92%|█████████▏| 13554/14725 [6:53:07<35:25,  1.82s/it]
 92%|█████████▏| 13555/14725 [6:53:09<35:02,  1.80s/it]
 92%|█████████▏| 13556/14725 [6:53:11<35:16,  1.81s/it]
 92%|█████████▏| 13557/14725 [6:53:12<34:55,  1.79s/it]
 92%|█████████▏| 13558/14725 [6:53:14<34:59,  1.80s/it]
 92%|█████████▏| 13559/14725 [6:53:16<34:43,  1.79s/it]
 92%|█████████▏| 13560/14725 [6:53:18<34:44,  1.79s/it]
                                                       

 92%|█████████▏| 13560/14725 [6:53:18<34:44,  1.79s/it]
 92%|█████████▏| 13561/14725 [6:53:20<35:13,  1.82s/it]
 92%|█████████▏| 13562/14725 [6:53:22<35:15,  1.82s/it]
 92%|█████████▏| 13563/14725 [6:53:23<35:36,  1.84s/it]
 92%|█████████▏| 13564/14725 [6:53:25<3
+0: {'loss': 0.8027, 'grad_norm': 0.6562379353171897, 'learning_rate': 2e-05, 'epoch': 0.92}
+0: 5:40,  1.84s/it]
 92%|█████████▏| 13565/14725 [6:53:27<35:15,  1.82s/it]
 92%|█████████▏| 13566/14725 [6:53:29<35:08,  1.82s/it]
 92%|█████████▏| 13567/14725 [6:53:31<35:12,  1.82s/it]
 92%|█████████▏| 13568/14725 [6:53:33<35:08,  1.82s/it]
 92%|█████████▏| 13569/14725 [6:53:34<35:47,  1.86s/it]
 92%|█████████▏| 13570/14725 [6:53:36<35:55,  1.87s/it]
                                                       

 92%|█████████▏| 13570/14725 [6:53:36<35:55,  1.87s/it]
 92%|█████████▏| 13571/14725 [6:53:38<35:20,  1.84s/it]
 92%|█████████▏| 13572/14725 [6:53:40<35:10,  1.83s/it]
 92%|█████████▏| 13573/14725 [6:53:42<34:55,  1.82s/it]
 92%|█████████▏| 13574/14725 [6:53:44<34:47,  1.81s/it]
 92%|█████████▏| 13575/14725 [6:53:45<35:03,  1.83s/it]
 92%|█████████▏| 1
+0: {'loss': 0.7718, 'grad_norm': 0.6448232728690747, 'learning_rate': 2e-05, 'epoch': 0.92}
+0: 3576/14725 [6:53:47<34:42,  1.81s/it]
 92%|█████████▏| 13577/14725 [6:53:49<34:54,  1.82s/it]
 92%|██████���██▏| 13578/14725 [6:53:51<34:29,  1.80s/it]
 92%|█████████▏| 13579/14725 [6:53:53<34:34,  1.81s/it]
 92%|█████████▏| 13580/14725 [6:53:54<34:20,  1.80s/it]
                                                       

 92%|█████████▏| 13580/14725 [6:53:54<34:20,  1.80s/it]
 92%|█████████▏| 13581/14725 [6:53:56<35:04,  1.84s/it]
 92%|█████████▏| 13582/14725 [6:53:58<35:09,  1.85s/it]
 92%|█████████▏| 13583/14725 [6:54:00<34:50,  1.83s/it]
 92%|█████████▏| 13584/14725 [6:54:02<34:47,  1.83s/it]
 92%|█████████▏| 13585/14725 [6:54:04<34:23,  1.81s/it]
 92%|█████████▏| 13586/14725 [6:54:05<34:38,  1.82s/it]
 92%|█████████▏| 13587/14725 [6:54:07<34:47,  1.83s/it]
 92%|████
+0: {'loss': 0.7898, 'grad_norm': 0.6835831334681927, 'learning_rate': 2e-05, 'epoch': 0.92}
+0: █████▏| 13588/14725 [6:54:09<35:13,  1.86s/it]
 92%|█████████▏| 13589/14725 [6:54:11<35:12,  1.86s/it]
 92%|█████████▏| 13590/14725 [6:54:13<35:07,  1.86s/it]
                                                       

 92%|█████████▏| 13590/14725 [6:54:13<35:07,  1.86s/it]
 92%|█████████▏| 13591/14725 [6:54:15<35:14,  1.86s/it]
 92%|█████████▏| 13592/14725 [6:54:17<35:04,  1.86s/it]
 92%|█████████▏| 13593/14725 [6:54:18<34:41,  1.84s/it]
 92%|█████████▏| 13594/14725 [6:54:20<34:13,  1.82s/it]
 92%|█████████▏| 13595/14725 [6:54:22<34:04,  1.81s/it]
 92%|█████████▏| 13596/14725 [6:54:24<34:16,  1.82s/it]
 92%|█████████▏| 13597/14725 [6:54:26<33:59,  1.81s/it]
 92%|█████████▏| 13598/14725 [6:54:27<33:55,  1.81s/it]
 92%|█████████▏| 13599/14725 [6:54:29<34:15,  1.83s/
+0: {'loss': 0.7837, 'grad_norm': 0.6877986052478406, 'learning_rate': 2e-05, 'epoch': 0.92}
+0: it]
 92%|█████████▏| 13600/14725 [6:54:31<34:01,  1.82s/it]
                                                       

 92%|█████████▏| 13600/14725 [6:54:31<34:01,  1.82s/it]
 92%|█████████▏| 13601/14725 [6:54:33<34:06,  1.82s/it]
 92%|█████████▏| 13602/14725 [6:54:35<33:47,  1.81s/it]
 92%|█████████▏| 13603/14725 [6:54:36<33:40,  1.80s/it]
 92%|█████████▏| 13604/14725 [6:54:38<33:23,  1.79s/it]
 92%|█████████▏| 13605/14725 [6:54:40<33:36,  1.80s/it]
 92%|█████████▏| 13606/14725 [6:54:42<33:34,  1.80s/it]
 92%|█████████▏| 13607/14725 [6:54:44<33:44,  1.81s/it]
 92%|█████████▏| 13608/14725 [6:54:45<33:17,  1.79s/it]
 92%|█████████▏| 13609/14725 [6:54:47<33:27,  1.80s/it]
 92%|█████████▏| 13610/14725 [6:54:49<33:35,  1.81s/it]
                                                   
+0: {'loss': 0.7871, 'grad_norm': 0.7015274282372794, 'learning_rate': 2e-05, 'epoch': 0.92}
+0: {'loss': 0.799, 'grad_norm': 0.6926623029282634, 'learning_rate': 2e-05, 'epoch': 0.92}
+0:     

 92%|█████████▏| 13610/14725 [6:54:49<33:35,  1.81s/it]
 92%|█████████▏| 13611/14725 [6:54:51<33:56,  1.83s/it]
 92%|█████████▏| 13612/14725 [6:54:53<33:42,  1.82s/it]
 92%|█████████▏| 13613/14725 [6:54:55<33:46,  1.82s/it]
 92%|█████████▏| 13614/14725 [6:54:56<33:30,  1.81s/it]
 92%|█████████▏| 13615/14725 [6:54:58<33:06,  1.79s/it]
 92%|█████████▏| 13616/14725 [6:55:00<33:25,  1.81s/it]
 92%|█████████▏| 13617/14725 [6:55:02<33:08,  1.79s/it]
 92%|█████████▏| 13618/14725 [6:55:03<32:58,  1.79s/it]
 92%|█████████▏| 13619/14725 [6:55:05<33:02,  1.79s/it]
 92%|█████████▏| 13620/14725 [6:55:07<33:28,  1.82s/it]
                                                       

 92%|█████████▏| 13620/14725 [6:55:07<33:28,  1.82s/it]
 93%|█████████▎| 13621/14725 
+0: {'loss': 0.7973, 'grad_norm': 0.6772445244234746, 'learning_rate': 2e-05, 'epoch': 0.93}
+0: [6:55:09<33:16,  1.81s/it]
 93%|█████████▎| 13622/14725 [6:55:11<33:16,  1.81s/it]
 93%|█████████▎| 13623/14725 [6:55:13<32:57,  1.79s/it]
 93%|█████████▎| 13624/14725 [6:55:14<33:10,  1.81s/it]
 93%|█████████▎| 13625/14725 [6:55:16<32:51,  1.79s/it]
 93%|█████████▎| 13626/14725 [6:55:18<33:28,  1.83s/it]
 93%|█████████▎| 13627/14725 [6:55:20<33:17,  1.82s/it]
 93%|█████████▎| 13628/14725 [6:55:22<33:10,  1.81s/it]
 93%|█████████▎| 13629/14725 [6:55:23<33:04,  1.81s/it]
 93%|█████████▎| 13630/14725 [6:55:25<32:50,  1.80s/it]
                                                       

 93%|█████████▎| 13630/14725 [6:55:25<32:50,  1.80s/it]
 93%|█████████▎| 13631/14725 [6:55:27<32:35,  1.79s/it]
 93%|█████████▎| 13632/14725 [6:55:29<32:20,  1.78s/it]
 93%|███████�
+0: {'loss': 0.7805, 'grad_norm': 0.6544302616652332, 'learning_rate': 2e-05, 'epoch': 0.93}
+0: �█▎| 13633/14725 [6:55:30<32:17,  1.77s/it]
 93%|█████████▎| 13634/14725 [6:55:32<32:22,  1.78s/it]
 93%|█████████▎| 13635/14725 [6:55:34<32:24,  1.78s/it]
 93%|█████████▎| 13636/14725 [6:55:36<32:38,  1.80s/it]
 93%|█████████▎| 13637/14725 [6:55:38<32:37,  1.80s/it]
 93%|█████████▎| 13638/14725 [6:55:40<32:40,  1.80s/it]
 93%|█████████▎| 13639/14725 [6:55:41<32:35,  1.80s/it]
 93%|█████████▎| 13640/14725 [6:55:43<33:18,  1.84s/it]
                                                       

 93%|█████████▎| 13640/14725 [6:55:43<33:18,  1.84s/it]
 93%|█████████▎| 13641/14725 [6:55:45<33:22,  1.85s/it]
 93%|█████████▎| 13642/14725 [6:55:47<32:55,  1.82s/it]
 93%|█████████▎| 13643/14725 [6:55:49<32:40,  1.81s/it]
 93%|█████████▎| 13644/14725 [6:55:51<33:49,  1.88s/it]
 93%|�
+0: {'loss': 0.7839, 'grad_norm': 0.7099657429977906, 'learning_rate': 2e-05, 'epoch': 0.93}
+0: �████████▎| 13645/14725 [6:55:52<33:12,  1.85s/it]
 93%|█████████▎| 13646/14725 [6:55:54<32:50,  1.83s/it]
 93%|█████████▎| 13647/14725 [6:55:56<32:25,  1.80s/it]
 93%|█████████▎| 13648/14725 [6:55:58<32:09,  1.79s/it]
 93%|█████████▎| 13649/14725 [6:56:00<32:17,  1.80s/it]
 93%|█████████▎| 13650/14725 [6:56:01<32:10,  1.80s/it]
                                                       

 93%|█████████▎| 13650/14725 [6:56:01<32:10,  1.80s/it]
 93%|█████████▎| 13651/14725 [6:56:03<32:26,  1.81s/it]
 93%|█████████▎| 13652/14725 [6:56:05<32:19,  1.81s/it]
 93%|█████████▎| 13653/14725 [6:56:07<32:23,  1.81s/it]
 93%|█████████▎| 13654/14725 [6:56:09<32:13,  1.81s/it]
 93%|█████████▎| 13655/14725 [6:56:10<32:08,  1.80s/it]
 93%|█████████▎| 13656/14725 [6:56:12<32:1
+0: {'loss': 0.7918, 'grad_norm': 0.6588346427120189, 'learning_rate': 2e-05, 'epoch': 0.93}
+0: 4,  1.81s/it]
 93%|█████████▎| 13657/14725 [6:56:14<32:03,  1.80s/it]
 93%|█████████▎| 13658/14725 [6:56:16<32:08,  1.81s/it]
 93%|█████████▎| 13659/14725 [6:56:18<32:03,  1.80s/it]
 93%|█████████▎| 13660/14725 [6:56:19<31:47,  1.79s/it]
                                                       

 93%|█████████▎| 13660/14725 [6:56:19<31:47,  1.79s/it]
 93%|█████████▎| 13661/14725 [6:56:21<31:43,  1.79s/it]
 93%|█████████▎| 13662/14725 [6:56:23<31:53,  1.80s/it]
 93%|█████████▎| 13663/14725 [6:56:25<31:54,  1.80s/it]
 93%|█████████▎| 13664/14725 [6:56:27<31:57,  1.81s/it]
 93%|█████████▎| 13665/14725 [6:56:28<32:00,  1.81s/it]
 93%|█████████▎| 13666/14725 [6:56:30<31:47,  1.80s/it]
 93%|█████████▎| 13667/14725 [6:56:32<31:45,  1.80s/it]
 93%|█████████▎| 1366
+0: {'loss': 0.7842, 'grad_norm': 0.6907850000258274, 'learning_rate': 2e-05, 'epoch': 0.93}
+0: 8/14725 [6:56:34<31:48,  1.81s/it]
 93%|█████████▎| 13669/14725 [6:56:36<32:06,  1.82s/it]
 93%|█████████▎| 13670/14725 [6:56:38<31:53,  1.81s/it]
                                                       

 93%|█████████▎| 13670/14725 [6:56:38<31:53,  1.81s/it]
 93%|█████████▎| 13671/14725 [6:56:39<31:49,  1.81s/it]
 93%|█████████▎| 13672/14725 [6:56:41<31:36,  1.80s/it]
 93%|█████████▎| 13673/14725 [6:56:43<31:38,  1.81s/it]
 93%|█████████▎| 13674/14725 [6:56:45<31:28,  1.80s/it]
 93%|█████████▎| 13675/14725 [6:56:46<31:22,  1.79s/it]
 93%|█████████▎| 13676/14725 [6:56:48<31:20,  1.79s/it]
 93%|█████████▎| 13677/14725 [6:56:50<31:11,  1.79s/it]
 93%|█████████▎| 13678/14725 [6:56:52<31:04,  1.78s/it]
 93%|█████████▎| 13679/14725 [6:56:54<30:58,  1.78s/it]
 93%|█████
+0: {'loss': 0.7774, 'grad_norm': 0.6615029818136764, 'learning_rate': 2e-05, 'epoch': 0.93}
+0: {'loss': 0.7807, 'grad_norm': 0.7009303791904142, 'learning_rate': 2e-05, 'epoch': 0.93}
+0: ████▎| 13680/14725 [6:56:55<30:51,  1.77s/it]
                                                       

 93%|█████████▎| 13680/14725 [6:56:55<30:51,  1.77s/it]
 93%|█████████▎| 13681/14725 [6:56:57<30:57,  1.78s/it]
 93%|█████████▎| 13682/14725 [6:56:59<31:01,  1.78s/it]
 93%|█████████▎| 13683/14725 [6:57:01<31:11,  1.80s/it]
 93%|█████████▎| 13684/14725 [6:57:03<31:38,  1.82s/it]
 93%|█████████▎| 13685/14725 [6:57:04<31:19,  1.81s/it]
 93%|█████████▎| 13686/14725 [6:57:06<31:13,  1.80s/it]
 93%|█████████▎| 13687/14725 [6:57:08<31:09,  1.80s/it]
 93%|█████████▎| 13688/14725 [6:57:10<30:56,  1.79s/it]
 93%|█████████▎| 13689/14725 [6:57:12<31:21,  1.82s/it]
 93%|█████████▎| 13690/14725 [6:57:13<31:11,  1.81s/it]
                                                       

 93%|████�
+0: {'loss': 0.7859, 'grad_norm': 0.6504887359714943, 'learning_rate': 2e-05, 'epoch': 0.93}
+0: ��████▎| 13690/14725 [6:57:13<31:11,  1.81s/it]
 93%|█████████▎| 13691/14725 [6:57:15<31:23,  1.82s/it]
 93%|█████████▎| 13692/14725 [6:57:17<31:36,  1.84s/it]
 93%|█████████▎| 13693/14725 [6:57:19<31:19,  1.82s/it]
 93%|█████████▎| 13694/14725 [6:57:21<31:11,  1.82s/it]
 93%|█████████▎| 13695/14725 [6:57:23<30:57,  1.80s/it]
 93%|█████████▎| 13696/14725 [6:57:24<30:50,  1.80s/it]
 93%|█████████▎| 13697/14725 [6:57:26<30:38,  1.79s/it]
 93%|█████████▎| 13698/14725 [6:57:28<30:30,  1.78s/it]
 93%|█████████▎| 13699/14725 [6:57:30<30:34,  1.79s/it]
 93%|█████████▎| 13700/14725 [6:57:31<30:44,  1.80s/it]
                                                       

 93%|█████████▎| 13700/14725 [6:57:31<30:44,  1.80s/it]
 93%|█████████▎| 13701/14725 [6:57:33<31:02,  1.82s/i
+0: {'loss': 0.8156, 'grad_norm': 0.6940572719936509, 'learning_rate': 2e-05, 'epoch': 0.93}
+0: t]
 93%|█████████▎| 13702/14725 [6:57:35<30:48,  1.81s/it]
 93%|█████████▎| 13703/14725 [6:57:37<31:05,  1.83s/it]
 93%|█████████▎| 13704/14725 [6:57:39<31:00,  1.82s/it]
 93%|█████████▎| 13705/14725 [6:57:41<31:15,  1.84s/it]
 93%|█████████▎| 13706/14725 [6:57:42<30:50,  1.82s/it]
 93%|█████████▎| 13707/14725 [6:57:44<30:47,  1.81s/it]
 93%|█████████▎| 13708/14725 [6:57:46<30:32,  1.80s/it]
 93%|█████████▎| 13709/14725 [6:57:48<30:48,  1.82s/it]
 93%|█████████▎| 13710/14725 [6:57:50<30:48,  1.82s/it]
                                                       

 93%|█████████▎| 13710/14725 [6:57:50<30:48,  1.82s/it]
 93%|█████████▎| 13711/14725 [6:57:52<30:54,  1.83s/it]
 93%|█████████▎| 13712/14725 [6:57:53<30:42,  1.82s/it]
 93%|█████████▎| 13713/14725 [6:
+0: {'loss': 0.7886, 'grad_norm': 0.6740140772555888, 'learning_rate': 2e-05, 'epoch': 0.93}
+0: 57:55<31:04,  1.84s/it]
 93%|█████████▎| 13714/14725 [6:57:57<30:55,  1.84s/it]
 93%|█████████▎| 13715/14725 [6:57:59<30:48,  1.83s/it]
 93%|█████████▎| 13716/14725 [6:58:01<30:52,  1.84s/it]
 93%|█████████▎| 13717/14725 [6:58:03<30:27,  1.81s/it]
 93%|█████████▎| 13718/14725 [6:58:04<30:52,  1.84s/it]
 93%|█████████▎| 13719/14725 [6:58:06<30:21,  1.81s/it]
 93%|█████████▎| 13720/14725 [6:58:08<30:51,  1.84s/it]
                                                       

 93%|█████████▎| 13720/14725 [6:58:08<30:51,  1.84s/it]
 93%|█████████▎| 13721/14725 [6:58:10<30:28,  1.82s/it]
 93%|█████████▎| 13722/14725 [6:58:12<30:15,  1.81s/it]
 93%|█████████▎| 13723/14725 [6:58:13<29:59,  1.80s/it]
 93%|█████████▎| 13724/14725 [6:58:15<29:47,  1.79s/it]
 93%|████████�
+0: {'loss': 0.7781, 'grad_norm': 0.6733613641024805, 'learning_rate': 2e-05, 'epoch': 0.93}
+0: �▎| 13725/14725 [6:58:17<29:47,  1.79s/it]
 93%|█████████▎| 13726/14725 [6:58:19<29:49,  1.79s/it]
 93%|█████████▎| 13727/14725 [6:58:21<29:50,  1.79s/it]
 93%|█████████▎| 13728/14725 [6:58:22<29:50,  1.80s/it]
 93%|█████████▎| 13729/14725 [6:58:24<29:37,  1.78s/it]
 93%|█████████▎| 13730/14725 [6:58:26<29:34,  1.78s/it]
                                                       

 93%|█████████▎| 13730/14725 [6:58:26<29:34,  1.78s/it]
 93%|█████████▎| 13731/14725 [6:58:28<29:49,  1.80s/it]
 93%|█████████▎| 13732/14725 [6:58:30<30:13,  1.83s/it]
 93%|█████████▎| 13733/14725 [6:58:31<30:27,  1.84s/it]
 93%|█████████▎| 13734/14725 [6:58:33<30:05,  1.82s/it]
 93%|█████████▎| 13735/14725 [6:58:35<29:48,  1.81s/it]
 93%|█████████▎| 13736/14725 [6:58:37<29:43,  1.80s/it]
 93%|█�
+0: {'loss': 0.7683, 'grad_norm': 0.6589021973260183, 'learning_rate': 2e-05, 'epoch': 0.93}
+0: �███████▎| 13737/14725 [6:58:39<29:57,  1.82s/it]
 93%|█████████▎| 13738/14725 [6:58:41<29:59,  1.82s/it]
 93%|█████████▎| 13739/14725 [6:58:42<29:55,  1.82s/it]
 93%|█████████▎| 13740/14725 [6:58:44<29:43,  1.81s/it]
                                                       

 93%|█████████▎| 13740/14725 [6:58:44<29:43,  1.81s/it]
 93%|█████████▎| 13741/14725 [6:58:46<31:11,  1.90s/it]
 93%|█████████▎| 13742/14725 [6:58:48<30:52,  1.88s/it]
 93%|█████████▎| 13743/14725 [6:58:50<30:26,  1.86s/it]
 93%|█████████▎| 13744/14725 [6:58:52<29:58,  1.83s/it]
 93%|█████████▎| 13745/14725 [6:58:53<29:52,  1.83s/it]
 93%|█████████▎| 13746/14725 [6:58:55<29:33,  1.81s/it]
 93%|█████████▎| 13747/14725 [6:58:57<30:16,  1.86s/it]
 93%|█████████▎| 13748/14725 [6:58:59<30:05, 
+0: {'loss': 0.8008, 'grad_norm': 0.6745008518049741, 'learning_rate': 2e-05, 'epoch': 0.93}
+0:  1.85s/it]
 93%|█████████▎| 13749/14725 [6:59:01<29:56,  1.84s/it]
 93%|█████████▎| 13750/14725 [6:59:03<31:03,  1.91s/it]
                                                       

 93%|█████████▎| 13750/14725 [6:59:03<31:03,  1.91s/it]
 93%|█████████▎| 13751/14725 [6:59:05<30:19,  1.87s/it]
 93%|█████████▎| 13752/14725 [6:59:07<30:30,  1.88s/it]
 93%|█████████▎| 13753/14725 [6:59:08<30:12,  1.87s/it]
 93%|█████████▎| 13754/14725 [6:59:10<29:36,  1.83s/it]
 93%|█████████▎| 13755/14725 [6:59:12<29:33,  1.83s/it]
 93%|█████████▎| 13756/14725 [6:59:14<29:15,  1.81s/it]
 93%|█████████▎| 13757/14725 [6:59:16<29:13,  1.81s/it]
 93%|█████████▎| 13758/14725 [6:59:17<29:31,  1.83s/it]
 93%|█████████▎| 13759/14725 [6:59:19<29:09,  1.81s/it]
 93%|█████████▎| 13760/1
+0: {'loss': 0.7828, 'grad_norm': 0.6519061718458458, 'learning_rate': 2e-05, 'epoch': 0.93}
+0: {'loss': 0.7855, 'grad_norm': 0.7027228523375483, 'learning_rate': 2e-05, 'epoch': 0.94}
+0: 4725 [6:59:21<28:55,  1.80s/it]
                                                       

 93%|█████████▎| 13760/14725 [6:59:21<28:55,  1.80s/it]
 93%|█████████▎| 13761/14725 [6:59:23<28:58,  1.80s/it]
 93%|█████████▎| 13762/14725 [6:59:25<29:09,  1.82s/it]
 93%|█████████▎| 13763/14725 [6:59:27<29:14,  1.82s/it]
 93%|█████████▎| 13764/14725 [6:59:28<29:19,  1.83s/it]
 93%|█████████▎| 13765/14725 [6:59:30<29:19,  1.83s/it]
 93%|█████████▎| 13766/14725 [6:59:32<29:15,  1.83s/it]
 93%|█████████▎| 13767/14725 [6:59:34<28:57,  1.81s/it]
 94%|█████████▎| 13768/14725 [6:59:36<28:48,  1.81s/it]
 94%|█████████▎| 13769/14725 [6:59:37<28:55,  1.82s/it]
 94%|█████████▎| 13770/14725 [6:59:39<28:48,  1.81s/it]
                                                       

 94%|█████████▎| 13770
+0: {'loss': 0.7831, 'grad_norm': 0.7073919646168064, 'learning_rate': 2e-05, 'epoch': 0.94}
+0: /14725 [6:59:39<28:48,  1.81s/it]
 94%|█████████▎| 13771/14725 [6:59:41<28:37,  1.80s/it]
 94%|█████████▎| 13772/14725 [6:59:43<28:39,  1.80s/it]
 94%|█████████▎| 13773/14725 [6:59:45<28:22,  1.79s/it]
 94%|█████████▎| 13774/14725 [6:59:46<28:25,  1.79s/it]
 94%|█████████▎| 13775/14725 [6:59:48<28:12,  1.78s/it]
 94%|█████████▎| 13776/14725 [6:59:50<28:20,  1.79s/it]
 94%|█████████▎| 13777/14725 [6:59:52<28:07,  1.78s/it]
 94%|█████████▎| 13778/14725 [6:59:54<28:24,  1.80s/it]
 94%|█████████▎| 13779/14725 [6:59:55<28:16,  1.79s/it]
 94%|█████████▎| 13780/14725 [6:59:57<28:18,  1.80s/it]
                                                       

 94%|█████████▎| 13780/14725 [6:59:57<28:18,  1.80s/it]
 94%|█████████▎| 13781/14725 [6:59:59<28:23,  1.80s/it]
 94%|█████�
+0: {'loss': 0.7876, 'grad_norm': 0.6522788174478849, 'learning_rate': 2e-05, 'epoch': 0.94}
+0: ��███▎| 13782/14725 [7:00:01<28:21,  1.80s/it]
 94%|█████████▎| 13783/14725 [7:00:03<28:41,  1.83s/it]
 94%|██��██████▎| 13784/14725 [7:00:05<29:03,  1.85s/it]
 94%|█████████▎| 13785/14725 [7:00:06<28:32,  1.82s/it]
 94%|█████████▎| 13786/14725 [7:00:08<28:31,  1.82s/it]
 94%|█████████▎| 13787/14725 [7:00:10<28:09,  1.80s/it]
 94%|█████████▎| 13788/14725 [7:00:12<28:05,  1.80s/it]
 94%|█████████▎| 13789/14725 [7:00:13<28:10,  1.81s/it]
 94%|█████████▎| 13790/14725 [7:00:15<28:06,  1.80s/it]
                                                       

 94%|█████████▎| 13790/14725 [7:00:15<28:06,  1.80s/it]
 94%|█████████▎| 13791/14725 [7:00:17<28:15,  1.82s/it]
 94%|█████████▎| 13792/14725 [7:00:19<28:09,  1.81s/it]
 94%|█████████▎| 13793/14725 [7:00:21<28:07,  1.81s/it]
+0: {'loss': 0.7918, 'grad_norm': 0.6680772180916851, 'learning_rate': 2e-05, 'epoch': 0.94}
+0:  94%|█████████▎| 13794/14725 [7:00:22<27:48,  1.79s/it]
 94%|█████████▎| 13795/14725 [7:00:24<28:09,  1.82s/it]
 94%|█████████▎| 13796/14725 [7:00:26<28:02,  1.81s/it]
 94%|█████████▎| 13797/14725 [7:00:28<27:56,  1.81s/it]
 94%|█████████▎| 13798/14725 [7:00:30<27:58,  1.81s/it]
 94%|█████████▎| 13799/14725 [7:00:32<28:05,  1.82s/it]
 94%|█████████▎| 13800/14725 [7:00:33<28:00,  1.82s/it]
                                                       

 94%|█████████▎| 13800/14725 [7:00:33<28:00,  1.82s/it]
 94%|█████████▎| 13801/14725 [7:00:35<27:47,  1.80s/it]
 94%|█████████▎| 13802/14725 [7:00:37<27:56,  1.82s/it]
 94%|█████████▎| 13803/14725 [7:00:39<27:56,  1.82s/it]
 94%|█████████▎| 13804/14725 [7:00:41<27:57,  1.82s/it]
 94%|█████████▍| 13805/14725 [7:00:
+0: {'loss': 0.8057, 'grad_norm': 0.6639392602482109, 'learning_rate': 2e-05, 'epoch': 0.94}
+0: 42<27:54,  1.82s/it]
 94%|█████████▍| 13806/14725 [7:00:44<27:58,  1.83s/it]
 94%|█████████▍| 13807/14725 [7:00:46<27:57,  1.83s/it]
 94%|█████████▍| 13808/14725 [7:00:48<28:04,  1.84s/it]
 94%|█████████▍| 13809/14725 [7:00:50<28:08,  1.84s/it]
 94%|█████████▍| 13810/14725 [7:00:52<27:48,  1.82s/it]
                                                       

 94%|█████████▍| 13810/14725 [7:00:52<27:48,  1.82s/it]
 94%|█████████▍| 13811/14725 [7:00:53<27:39,  1.82s/it]
 94%|█████████▍| 13812/14725 [7:00:55<27:26,  1.80s/it]
 94%|█████████▍| 13813/14725 [7:00:57<27:15,  1.79s/it]
 94%|█████████▍| 13814/14725 [7:00:59<27:10,  1.79s/it]
 94%|█████████▍| 13815/14725 [7:01:01<27:22,  1.80s/it]
 94%|█████████▍| 13816/14725 [7:01:03<27:49,  1.84s/it]
 94%|█████████�
+0: {'loss': 0.7778, 'grad_norm': 0.7163757118290927, 'learning_rate': 2e-05, 'epoch': 0.94}
+0: �| 13817/14725 [7:01:04<27:31,  1.82s/it]
 94%|█████████▍| 13818/14725 [7:01:06<27:23,  1.81s/it]
 94%|█████████▍| 13819/14725 [7:01:08<27:28,  1.82s/it]
 94%|█████████▍| 13820/14725 [7:01:10<27:30,  1.82s/it]
                                                       

 94%|█████████▍| 13820/14725 [7:01:10<27:30,  1.82s/it]
 94%|█████████▍| 13821/14725 [7:01:12<27:23,  1.82s/it]
 94%|█████████▍| 13822/14725 [7:01:13<27:14,  1.81s/it]
 94%|█████████▍| 13823/14725 [7:01:15<27:10,  1.81s/it]
 94%|█████████▍| 13824/14725 [7:01:17<27:05,  1.80s/it]
 94%|█████████▍| 13825/14725 [7:01:19<26:51,  1.79s/it]
 94%|█████████▍| 13826/14725 [7:01:21<27:05,  1.81s/it]
 94%|█████████▍| 13827/14725 [7:01:22<26:58,  1.80s/it]
 94%|█████████▍| 13828/14725 [7:01:24<26:55,  1.80s/it]
 94%|██�
+0: {'loss': 0.7924, 'grad_norm': 0.726248341490536, 'learning_rate': 2e-05, 'epoch': 0.94}
+0: �██████▍| 13829/14725 [7:01:26<26:46,  1.79s/it]
 94%|█████████▍| 13830/14725 [7:01:28<26:36,  1.78s/it]
                                                       

 94%|█████████▍| 13830/14725 [7:01:28<26:36,  1.78s/it]
 94%|█████████▍| 13831/14725 [7:01:30<27:12,  1.83s/it]
 94%|█████████▍| 13832/14725 [7:01:31<27:04,  1.82s/it]
 94%|█████████▍| 13833/14725 [7:01:33<26:54,  1.81s/it]
 94%|█████████▍| 13834/14725 [7:01:35<26:45,  1.80s/it]
 94%|█████████▍| 13835/14725 [7:01:37<27:05,  1.83s/it]
 94%|█████████▍| 13836/14725 [7:01:39<27:38,  1.87s/it]
 94%|█████████▍| 13837/14725 [7:01:41<27:22,  1.85s/it]
 94%|█████████▍| 13838/14725 [7:01:42<26:56,  1.82s/it]
 94%|█████████▍| 13839/14725 [7:01:44<27:32,  1.86s/it]
 94%|█████████▍| 13840/14725 [7:01:46<27:08,  1.
+0: {'loss': 0.7715, 'grad_norm': 0.5914156716130946, 'learning_rate': 2e-05, 'epoch': 0.94}
+0: {'loss': 0.7959, 'grad_norm': 0.6644276228506384, 'learning_rate': 2e-05, 'epoch': 0.94}
+0: 84s/it]
                                                       

 94%|█████████▍| 13840/14725 [7:01:46<27:08,  1.84s/it]
 94%|█████████▍| 13841/14725 [7:01:48<26:54,  1.83s/it]
 94%|█████████▍| 13842/14725 [7:01:50<26:31,  1.80s/it]
 94%|█████████▍| 13843/14725 [7:01:52<26:38,  1.81s/it]
 94%|█████████▍| 13844/14725 [7:01:53<26:35,  1.81s/it]
 94%|█████████▍| 13845/14725 [7:01:55<26:49,  1.83s/it]
 94%|█████████▍| 13846/14725 [7:01:57<26:25,  1.80s/it]
 94%|█████████▍| 13847/14725 [7:01:59<26:36,  1.82s/it]
 94%|█████████▍| 13848/14725 [7:02:01<26:37,  1.82s/it]
 94%|█████████▍| 13849/14725 [7:02:02<26:17,  1.80s/it]
 94%|█████████▍| 13850/14725 [7:02:04<26:10,  1.79s/it]
                                                       

 94%|█████████▍| 13850/14725 [7:02:04<26:10,  
+0: {'loss': 0.7701, 'grad_norm': 0.6603080277656321, 'learning_rate': 2e-05, 'epoch': 0.94}
+0: 1.79s/it]
 94%|█████████▍| 13851/14725 [7:02:06<26:18,  1.81s/it]
 94%|█████████▍| 13852/14725 [7:02:08<26:22,  1.81s/it]
 94%|█████████▍| 13853/14725 [7:02:10<26:26,  1.82s/it]
 94%|█████████▍| 13854/14725 [7:02:11<26:17,  1.81s/it]
 94%|█████████▍| 13855/14725 [7:02:13<26:07,  1.80s/it]
 94%|█████████▍| 13856/14725 [7:02:15<26:22,  1.82s/it]
 94%|█████████▍| 13857/14725 [7:02:17<26:26,  1.83s/it]
 94%|█████████▍| 13858/14725 [7:02:19<26:20,  1.82s/it]
 94%|█████████▍| 13859/14725 [7:02:21<26:02,  1.80s/it]
 94%|█████████▍| 13860/14725 [7:02:22<25:56,  1.80s/it]
                                                       

 94%|█████████▍| 13860/14725 [7:02:22<25:56,  1.80s/it]
 94%|█████████▍| 13861/14725 [7:02:24<25:51,  1.80s/it]
 94%|█████████▍| 13862/14
+0: {'loss': 0.7763, 'grad_norm': 0.6806477746697949, 'learning_rate': 2e-05, 'epoch': 0.94}
+0: 725 [7:02:26<25:39,  1.78s/it]
 94%|█████████▍| 13863/14725 [7:02:28<25:55,  1.80s/it]
 94%|█████████▍| 13864/14725 [7:02:29<25:39,  1.79s/it]
 94%|█████████▍| 13865/14725 [7:02:31<25:59,  1.81s/it]
 94%|█████████▍| 13866/14725 [7:02:33<25:55,  1.81s/it]
 94%|█████████▍| 13867/14725 [7:02:35<25:57,  1.82s/it]
 94%|█████████▍| 13868/14725 [7:02:37<25:54,  1.81s/it]
 94%|█████████▍| 13869/14725 [7:02:39<26:06,  1.83s/it]
 94%|█████████▍| 13870/14725 [7:02:41<26:16,  1.84s/it]
                                                       

 94%|█████████▍| 13870/14725 [7:02:41<26:16,  1.84s/it]
 94%|█████████▍| 13871/14725 [7:02:42<25:58,  1.82s/it]
 94%|█████████▍| 13872/14725 [7:02:44<25:37,  1.80s/it]
 94%|█████████▍| 13873/14725 [7:02:46<25:36,  1.80s/it]
 94%|██████�
+0: {'loss': 0.7744, 'grad_norm': 0.6866291479545372, 'learning_rate': 2e-05, 'epoch': 0.94}
+0: ��██▍| 13874/14725 [7:02:48<25:27,  1.79s/it]
 94%|█████████▍| 13875/14725 [7:02:50<25:45,  1.82s/it]
 94%|█████████▍| 13876/14725 [7:02:51<25:31,  1.80s/it]
 94%|█████████▍| 13877/14725 [7:02:53<25:17,  1.79s/it]
 94%|█████████▍| 13878/14725 [7:02:55<25:23,  1.80s/it]
 94%|█████████▍| 13879/14725 [7:02:57<25:29,  1.81s/it]
 94%|█████████▍| 13880/14725 [7:02:58<25:21,  1.80s/it]
                                                       

 94%|█████████▍| 13880/14725 [7:02:58<25:21,  1.80s/it]
 94%|█████████▍| 13881/14725 [7:03:00<25:12,  1.79s/it]
 94%|█████████▍| 13882/14725 [7:03:02<25:33,  1.82s/it]
 94%|█████████▍| 13883/14725 [7:03:04<25:11,  1.79s/it]
 94%|█████████▍| 13884/14725 [7:03:06<25:24,  1.81s/it]
 94%|█████████▍| 13885/14725 [7:03:08<25:18,  1.81s/it]
 94
+0: {'loss': 0.8048, 'grad_norm': 0.7149506671068692, 'learning_rate': 2e-05, 'epoch': 0.94}
+0: %|█████████▍| 13886/14725 [7:03:09<25:13,  1.80s/it]
 94%|█████████▍| 13887/14725 [7:03:11<25:21,  1.82s/it]
 94%|█████████▍| 13888/14725 [7:03:13<25:08,  1.80s/it]
 94%|█████████▍| 13889/14725 [7:03:15<25:14,  1.81s/it]
 94%|█████████▍| 13890/14725 [7:03:17<25:14,  1.81s/it]
                                                       

 94%|█████████▍| 13890/14725 [7:03:17<25:14,  1.81s/it]
 94%|█████████▍| 13891/14725 [7:03:18<25:01,  1.80s/it]
 94%|█████████▍| 13892/14725 [7:03:20<25:19,  1.82s/it]
 94%|█████████▍| 13893/14725 [7:03:22<25:08,  1.81s/it]
 94%|█████████▍| 13894/14725 [7:03:24<25:06,  1.81s/it]
 94%|█████████▍| 13895/14725 [7:03:26<25:13,  1.82s/it]
 94%|█████████▍| 13896/14725 [7:03:27<25:00,  1.81s/it]
 94%|█████████▍| 13897/14725 [7:03:29<
+0: {'loss': 0.8054, 'grad_norm': 0.6599428695901804, 'learning_rate': 2e-05, 'epoch': 0.94}
+0: 25:07,  1.82s/it]
 94%|█████████▍| 13898/14725 [7:03:31<25:13,  1.83s/it]
 94%|█████████▍| 13899/14725 [7:03:33<25:41,  1.87s/it]
 94%|█████████▍| 13900/14725 [7:03:35<25:49,  1.88s/it]
                                                       

 94%|█████████▍| 13900/14725 [7:03:35<25:49,  1.88s/it]
 94%|█████████▍| 13901/14725 [7:03:37<25:26,  1.85s/it]
 94%|█████████▍| 13902/14725 [7:03:39<25:31,  1.86s/it]
 94%|█████████▍| 13903/14725 [7:03:40<25:14,  1.84s/it]
 94%|█████████▍| 13904/14725 [7:03:42<25:12,  1.84s/it]
 94%|█████████▍| 13905/14725 [7:03:44<25:07,  1.84s/it]
 94%|█████████▍| 13906/14725 [7:03:46<24:43,  1.81s/it]
 94%|█████████▍| 13907/14725 [7:03:48<24:50,  1.82s/it]
 94%|█████████▍| 13908/14725 [7:03:50<24:41,  1.81s/it]
 94%|█████████▍| 
+0: {'loss': 0.7945, 'grad_norm': 0.6510915482097706, 'learning_rate': 2e-05, 'epoch': 0.94}
+0: 13909/14725 [7:03:51<24:47,  1.82s/it]
 94%|█████████▍| 13910/14725 [7:03:53<25:04,  1.85s/it]
                                                       

 94%|█████████▍| 13910/14725 [7:03:53<25:04,  1.85s/it]
 94%|█████████▍| 13911/14725 [7:03:55<25:01,  1.84s/it]
 94%|█████████▍| 13912/14725 [7:03:57<25:02,  1.85s/it]
 94%|█████████▍| 13913/14725 [7:03:59<24:47,  1.83s/it]
 94%|█████████▍| 13914/14725 [7:04:01<24:28,  1.81s/it]
 94%|█████████▍| 13915/14725 [7:04:02<24:45,  1.83s/it]
 95%|█████████▍| 13916/14725 [7:04:04<24:23,  1.81s/it]
 95%|█████████▍| 13917/14725 [7:04:06<24:47,  1.84s/it]
 95%|█████████▍| 13918/14725 [7:04:08<24:55,  1.85s/it]
 95%|█████████▍| 13919/14725 [7:04:10<24:29,  1.82s/it]
 95%|█████████▍| 13920/14725 [7:04:11<24:11,  1.80s/it]
                
+0: {'loss': 0.7956, 'grad_norm': 0.660826409966034, 'learning_rate': 2e-05, 'epoch': 0.95}
+0: {'loss': 0.7807, 'grad_norm': 0.6699875840331447, 'learning_rate': 2e-05, 'epoch': 0.95}
+0:                                        

 95%|█████████▍| 13920/14725 [7:04:11<24:11,  1.80s/it]
 95%|█████████▍| 13921/14725 [7:04:13<24:11,  1.81s/it]
 95%|█████████▍| 13922/14725 [7:04:15<24:24,  1.82s/it]
 95%|█████████▍| 13923/14725 [7:04:17<25:58,  1.94s/it]
 95%|█████████▍| 13924/14725 [7:04:19<25:15,  1.89s/it]
 95%|█████████▍| 13925/14725 [7:04:21<24:39,  1.85s/it]
 95%|█████████▍| 13926/14725 [7:04:23<24:18,  1.83s/it]
 95%|█████████▍| 13927/14725 [7:04:25<25:55,  1.95s/it]
 95%|█████████▍| 13928/14725 [7:04:27<25:11,  1.90s/it]
 95%|█████████▍| 13929/14725 [7:04:28<24:34,  1.85s/it]
 95%|█████████▍| 13930/14725 [7:04:30<24:07,  1.82s/it]
                                                       

 95%|█████████▍| 13930/14725 [7:04:30<24:07,  1.82s/it]
 95%|███
+0: {'loss': 0.8107, 'grad_norm': 0.6894548429179759, 'learning_rate': 2e-05, 'epoch': 0.95}
+0: ██████▍| 13931/14725 [7:04:32<23:58,  1.81s/it]
 95%|█████████▍| 13932/14725 [7:04:34<23:58,  1.81s/it]
 95%|█████████▍| 13933/14725 [7:04:36<23:40,  1.79s/it]
 95%|█████████▍| 13934/14725 [7:04:37<24:06,  1.83s/it]
 95%|█████████▍| 13935/14725 [7:04:39<23:46,  1.81s/it]
 95%|█████████▍| 13936/14725 [7:04:41<23:43,  1.80s/it]
 95%|█████████▍| 13937/14725 [7:04:43<23:50,  1.82s/it]
 95%|█████████▍| 13938/14725 [7:04:45<23:50,  1.82s/it]
 95%|█████████▍| 13939/14725 [7:04:46<23:35,  1.80s/it]
 95%|█████████▍| 13940/14725 [7:04:48<23:34,  1.80s/it]
                                                       

 95%|█████████▍| 13940/14725 [7:04:48<23:34,  1.80s/it]
 95%|█████████▍| 13941/14725 [7:04:50<23:32,  1.80s/it]
 95%|█████████▍| 13942/14725 [7:04:52<23:35,  1.8
+0: {'loss': 0.802, 'grad_norm': 0.6236997592375502, 'learning_rate': 2e-05, 'epoch': 0.95}
+0: 1s/it]
 95%|█████████▍| 13943/14725 [7:04:54<23:36,  1.81s/it]
 95%|█████████▍| 13944/14725 [7:04:55<23:24,  1.80s/it]
 95%|█████████▍| 13945/14725 [7:04:57<23:15,  1.79s/it]
 95%|█████████▍| 13946/14725 [7:04:59<23:18,  1.79s/it]
 95%|█████████▍| 13947/14725 [7:05:01<23:25,  1.81s/it]
 95%|█████████▍| 13948/14725 [7:05:03<23:25,  1.81s/it]
 95%|█████████▍| 13949/14725 [7:05:05<23:41,  1.83s/it]
 95%|█████████▍| 13950/14725 [7:05:06<23:24,  1.81s/it]
                                                       

 95%|█████████▍| 13950/14725 [7:05:06<23:24,  1.81s/it]
 95%|█████████▍| 13951/14725 [7:05:08<23:18,  1.81s/it]
 95%|█████████▍| 13952/14725 [7:05:10<24:17,  1.89s/it]
 95%|█████████▍| 13953/14725 [7:05:12<24:00,  1.87s/it]
 95%|█████████▍| 13954/14725
+0: {'loss': 0.7933, 'grad_norm': 0.659065333385349, 'learning_rate': 2e-05, 'epoch': 0.95}
+0:  [7:05:14<24:14,  1.89s/it]
 95%|█████████▍| 13955/14725 [7:05:16<23:50,  1.86s/it]
 95%|█████████▍| 13956/14725 [7:05:18<23:43,  1.85s/it]
 95%|█████████▍| 13957/14725 [7:05:19<23:17,  1.82s/it]
 95%|█████████▍| 13958/14725 [7:05:21<23:12,  1.82s/it]
 95%|█████████▍| 13959/14725 [7:05:23<23:17,  1.82s/it]
 95%|█████████▍| 13960/14725 [7:05:25<23:07,  1.81s/it]
                                                       

 95%|█████████▍| 13960/14725 [7:05:25<23:07,  1.81s/it]
 95%|█████████▍| 13961/14725 [7:05:27<23:05,  1.81s/it]
 95%|█████████▍| 13962/14725 [7:05:28<22:55,  1.80s/it]
 95%|█████████▍| 13963/14725 [7:05:30<23:02,  1.81s/it]
 95%|█████████▍| 13964/14725 [7:05:32<23:00,  1.81s/it]
 95%|█████████▍| 13965/14725 [7:05:34<23:31,  1.86s/it]
 95%|███████�
+0: {'loss': 0.781, 'grad_norm': 0.697950708233614, 'learning_rate': 2e-05, 'epoch': 0.95}
+0: ��█▍| 13966/14725 [7:05:36<23:18,  1.84s/it]
 95%|█████████▍| 13967/14725 [7:05:38<23:18,  1.84s/it]
 95%|█████████▍| 13968/14725 [7:05:40<24:26,  1.94s/it]
 95%|█████████▍| 13969/14725 [7:05:42<23:50,  1.89s/it]
 95%|█████████▍| 13970/14725 [7:05:43<23:18,  1.85s/it]
                                                       

 95%|█████████▍| 13970/14725 [7:05:43<23:18,  1.85s/it]
 95%|█████████▍| 13971/14725 [7:05:45<23:08,  1.84s/it]
 95%|█████████▍| 13972/14725 [7:05:47<22:53,  1.82s/it]
 95%|█████████▍| 13973/14725 [7:05:49<22:38,  1.81s/it]
 95%|█████████▍| 13974/14725 [7:05:51<22:42,  1.81s/it]
 95%|█████████▍| 13975/14725 [7:05:52<22:39,  1.81s/it]
 95%|█████████▍| 13976/14725 [7:05:54<22:41,  1.82s/it]
 95%|█████████▍| 13977/14725 [7:05:56<22:32,  1.81s/it]
 95%|�
+0: {'loss': 0.7958, 'grad_norm': 0.6456316026195805, 'learning_rate': 2e-05, 'epoch': 0.95}
+0: ��████████▍| 13978/14725 [7:05:58<22:32,  1.81s/it]
 95%|█████████▍| 13979/14725 [7:06:00<22:25,  1.80s/it]
 95%|█████████▍| 13980/14725 [7:06:01<22:08,  1.78s/it]
                                                       

 95%|█████████▍| 13980/14725 [7:06:01<22:08,  1.78s/it]
 95%|█████████▍| 13981/14725 [7:06:03<22:23,  1.81s/it]
 95%|█████████▍| 13982/14725 [7:06:05<22:11,  1.79s/it]
 95%|█████████▍| 13983/14725 [7:06:07<22:16,  1.80s/it]
 95%|█████████▍| 13984/14725 [7:06:09<22:19,  1.81s/it]
 95%|█████████▍| 13985/14725 [7:06:10<22:15,  1.80s/it]
 95%|█████████▍| 13986/14725 [7:06:12<22:02,  1.79s/it]
 95%|█████████▍| 13987/14725 [7:06:14<22:17,  1.81s/it]
 95%|█████████▍| 13988/14725 [7:06:16<22:01,  1.79s/it]
 95%|█████████▌| 13989/14725 [7:06:17<21:
+0: {'loss': 0.8012, 'grad_norm': 0.6694768071387902, 'learning_rate': 2e-05, 'epoch': 0.95}
+0: 59,  1.79s/it]
 95%|█████████▌| 13990/14725 [7:06:19<22:12,  1.81s/it]
                                                       

 95%|█████████▌| 13990/14725 [7:06:19<22:12,  1.81s/it]
 95%|██████���██▌| 13991/14725 [7:06:21<22:14,  1.82s/it]
 95%|█████████▌| 13992/14725 [7:06:23<22:20,  1.83s/it]
 95%|█████████▌| 13993/14725 [7:06:25<22:13,  1.82s/it]
 95%|█████████▌| 13994/14725 [7:06:27<22:02,  1.81s/it]
 95%|█████████▌| 13995/14725 [7:06:28<22:03,  1.81s/it]
 95%|█████████▌| 13996/14725 [7:06:30<21:50,  1.80s/it]
 95%|█████████▌| 13997/14725 [7:06:32<21:47,  1.80s/it]
 95%|█████████▌| 13998/14725 [7:06:34<21:45,  1.80s/it]
 95%|█████████▌| 13999/14725 [7:06:36<21:32,  1.78s/it]
 95%|█████████▌| 14000/14725 [7:06:37<21:38,  1.79s/it]
                                        
+0: {'loss': 0.7838, 'grad_norm': 0.6328269277365495, 'learning_rate': 2e-05, 'epoch': 0.95}
+0: {'loss': 0.8023, 'grad_norm': 0.642770174986998, 'learning_rate': 2e-05, 'epoch': 0.95}
+0:                

 95%|█████████▌| 14000/14725 [7:06:37<21:38,  1.79s/it]
 95%|█████████▌| 14001/14725 [7:06:39<21:35,  1.79s/it]
 95%|█████████▌| 14002/14725 [7:06:41<21:30,  1.78s/it]
 95%|█████████▌| 14003/14725 [7:06:43<21:27,  1.78s/it]
 95%|█████████▌| 14004/14725 [7:06:44<21:31,  1.79s/it]
 95%|█████████▌| 14005/14725 [7:06:46<21:30,  1.79s/it]
 95%|█████████▌| 14006/14725 [7:06:48<21:32,  1.80s/it]
 95%|█████████▌| 14007/14725 [7:06:50<21:55,  1.83s/it]
 95%|█████████▌| 14008/14725 [7:06:52<21:48,  1.82s/it]
 95%|█████████▌| 14009/14725 [7:06:54<21:37,  1.81s/it]
 95%|█████████▌| 14010/14725 [7:06:55<21:28,  1.80s/it]
                                                       

 95%|█████████▌| 14010/14725 [7:06:55<21:28,  1.80s/it]
 95%|█████████▌| 1
+0: {'loss': 0.7968, 'grad_norm': 0.6430309039574226, 'learning_rate': 2e-05, 'epoch': 0.95}
+0: 4011/14725 [7:06:57<21:15,  1.79s/it]
 95%|█████████▌| 14012/14725 [7:06:59<21:22,  1.80s/it]
 95%|█████████▌| 14013/14725 [7:07:01<21:21,  1.80s/it]
 95%|█████████▌| 14014/14725 [7:07:03<21:20,  1.80s/it]
 95%|█████████▌| 14015/14725 [7:07:04<21:25,  1.81s/it]
 95%|█████████▌| 14016/14725 [7:07:06<21:24,  1.81s/it]
 95%|█████████▌| 14017/14725 [7:07:08<21:19,  1.81s/it]
 95%|█████████▌| 14018/14725 [7:07:10<21:20,  1.81s/it]
 95%|█████████▌| 14019/14725 [7:07:12<21:26,  1.82s/it]
 95%|█████████▌| 14020/14725 [7:07:13<21:13,  1.81s/it]
                                                       

 95%|█████████▌| 14020/14725 [7:07:13<21:13,  1.81s/it]
 95%|█████████▌| 14021/14725 [7:07:15<21:22,  1.82s/it]
 95%|█████████▌| 14022/14725 [7:07:17<21:14,  1.81s/it]
 95%|████
+0: {'loss': 0.7873, 'grad_norm': 0.6515630978270521, 'learning_rate': 2e-05, 'epoch': 0.95}
+0: █████▌| 14023/14725 [7:07:19<21:15,  1.82s/it]
 95%|█████████▌| 14024/14725 [7:07:21<21:05,  1.81s/it]
 95%|█████████▌| 14025/14725 [7:07:23<21:22,  1.83s/it]
 95%|█████████▌| 14026/14725 [7:07:24<21:00,  1.80s/it]
 95%|█████████▌| 14027/14725 [7:07:26<20:52,  1.79s/it]
 95%|█████████▌| 14028/14725 [7:07:28<20:51,  1.80s/it]
 95%|█████████▌| 14029/14725 [7:07:30<20:54,  1.80s/it]
 95%|█████████▌| 14030/14725 [7:07:32<21:00,  1.81s/it]
                                                       

 95%|█████████▌| 14030/14725 [7:07:32<21:00,  1.81s/it]
 95%|█████████▌| 14031/14725 [7:07:33<21:02,  1.82s/it]
 95%|█████████▌| 14032/14725 [7:07:35<21:00,  1.82s/it]
 95%|█████████▌| 14033/14725 [7:07:37<21:15,  1.84s/it]
 95%|█████████▌| 14034/14725 [7:07:39<21:00,  1.82s/
+0: {'loss': 0.7878, 'grad_norm': 0.6567395340593646, 'learning_rate': 2e-05, 'epoch': 0.95}
+0: it]
 95%|█████████▌| 14035/14725 [7:07:41<20:49,  1.81s/it]
 95%|█████████▌| 14036/14725 [7:07:42<20:48,  1.81s/it]
 95%|█████████▌| 14037/14725 [7:07:44<20:45,  1.81s/it]
 95%|█████████▌| 14038/14725 [7:07:46<20:33,  1.80s/it]
 95%|█████████▌| 14039/14725 [7:07:48<20:34,  1.80s/it]
 95%|█████████▌| 14040/14725 [7:07:50<20:43,  1.81s/it]
                                                       

 95%|█████████▌| 14040/14725 [7:07:50<20:43,  1.81s/it]
 95%|█████████▌| 14041/14725 [7:07:51<20:32,  1.80s/it]
 95%|█████████▌| 14042/14725 [7:07:53<20:35,  1.81s/it]
 95%|█████████▌| 14043/14725 [7:07:55<20:23,  1.79s/it]
 95%|█████████▌| 14044/14725 [7:07:57<20:17,  1.79s/it]
 95%|█████████▌| 14045/14725 [7:07:59<20:13,  1.78s/it]
 95%|█████████▌| 14046/14725 [7
+0: {'loss': 0.7734, 'grad_norm': 0.663904457038954, 'learning_rate': 2e-05, 'epoch': 0.95}
+0: :08:00<20:11,  1.78s/it]
 95%|█████████▌| 14047/14725 [7:08:02<20:07,  1.78s/it]
 95%|█████████▌| 14048/14725 [7:08:04<20:02,  1.78s/it]
 95%|█████████▌| 14049/14725 [7:08:06<20:17,  1.80s/it]
 95%|█████████▌| 14050/14725 [7:08:08<20:31,  1.82s/it]
                                                       

 95%|█████████▌| 14050/14725 [7:08:08<20:31,  1.82s/it]
 95%|█████████▌| 14051/14725 [7:08:10<20:47,  1.85s/it]
 95%|█████████▌| 14052/14725 [7:08:11<20:31,  1.83s/it]
 95%|█████████▌| 14053/14725 [7:08:13<20:21,  1.82s/it]
 95%|█████████▌| 14054/14725 [7:08:15<20:12,  1.81s/it]
 95%|█████████▌| 14055/14725 [7:08:17<20:04,  1.80s/it]
 95%|█████████▌| 14056/14725 [7:08:18<19:57,  1.79s/it]
 95%|█████████▌| 14057/14725 [7:08:20<20:05,  1.80s/it]
 95%|████████�
+0: {'loss': 0.8183, 'grad_norm': 0.6856097457058328, 'learning_rate': 2e-05, 'epoch': 0.95}
+0: ��▌| 14058/14725 [7:08:22<20:02,  1.80s/it]
 95%|█████████▌| 14059/14725 [7:08:24<20:07,  1.81s/it]
 95%|█████████▌| 14060/14725 [7:08:26<20:00,  1.80s/it]
                                                       

 95%|█████████▌| 14060/14725 [7:08:26<20:00,  1.80s/it]
 95%|█████████▌| 14061/14725 [7:08:28<20:04,  1.81s/it]
 95%|█████████▌| 14062/14725 [7:08:29<19:57,  1.81s/it]
 96%|█████████▌| 14063/14725 [7:08:31<19:55,  1.81s/it]
 96%|█████████▌| 14064/14725 [7:08:33<19:50,  1.80s/it]
 96%|█████████▌| 14065/14725 [7:08:35<19:37,  1.78s/it]
 96%|█████████▌| 14066/14725 [7:08:37<19:48,  1.80s/it]
 96%|█████████▌| 14067/14725 [7:08:38<19:42,  1.80s/it]
 96%|█████████▌| 14068/14725 [7:08:40<19:54,  1.82s/it]
 96%|█████████▌| 14069/14725 [7:08:42<19:57,  1.83s/it]
 96%|█�
+0: {'loss': 0.7788, 'grad_norm': 0.718130207725363, 'learning_rate': 2e-05, 'epoch': 0.96}
+0: {'loss': 0.7963, 'grad_norm': 0.6622880796916327, 'learning_rate': 2e-05, 'epoch': 0.96}
+0: ��███████▌| 14070/14725 [7:08:44<19:47,  1.81s/it]
                                                       

 96%|█████████▌| 14070/14725 [7:08:44<19:47,  1.81s/it]
 96%|█████████▌| 14071/14725 [7:08:46<19:35,  1.80s/it]
 96%|█████████▌| 14072/14725 [7:08:47<19:38,  1.80s/it]
 96%|█████████▌| 14073/14725 [7:08:49<19:51,  1.83s/it]
 96%|█████████▌| 14074/14725 [7:08:51<19:42,  1.82s/it]
 96%|█████████▌| 14075/14725 [7:08:53<19:28,  1.80s/it]
 96%|█████████▌| 14076/14725 [7:08:55<19:26,  1.80s/it]
 96%|█████████▌| 14077/14725 [7:08:56<19:29,  1.80s/it]
 96%|█████████▌| 14078/14725 [7:08:58<19:22,  1.80s/it]
 96%|█████████▌| 14079/14725 [7:09:00<19:24,  1.80s/it]
 96%|█████████▌| 14080/14725 [7:09:02<19:14,  1.79s/it]
                                                       

 96%|�
+0: {'loss': 0.7833, 'grad_norm': 0.6569837872680981, 'learning_rate': 2e-05, 'epoch': 0.96}
+0: �████████▌| 14080/14725 [7:09:02<19:14,  1.79s/it]
 96%|█████████▌| 14081/14725 [7:09:04<19:26,  1.81s/it]
 96%|█████████▌| 14082/14725 [7:09:05<19:13,  1.79s/it]
 96%|█████████▌| 14083/14725 [7:09:07<19:20,  1.81s/it]
 96%|█████████▌| 14084/14725 [7:09:09<19:21,  1.81s/it]
 96%|█████████▌| 14085/14725 [7:09:11<19:21,  1.81s/it]
 96%|█████████▌| 14086/14725 [7:09:13<19:18,  1.81s/it]
 96%|█████████▌| 14087/14725 [7:09:15<19:21,  1.82s/it]
 96%|█████████▌| 14088/14725 [7:09:16<19:15,  1.81s/it]
 96%|█████████▌| 14089/14725 [7:09:18<19:11,  1.81s/it]
 96%|█████████▌| 14090/14725 [7:09:20<19:07,  1.81s/it]
                                                       

 96%|█████████▌| 14090/14725 [7:09:20<19:07,  1.81s/it]
 96%|█████████▌| 14091/14725 [7:09:22<19:0
+0: {'loss': 0.7907, 'grad_norm': 0.6263633576705125, 'learning_rate': 2e-05, 'epoch': 0.96}
+0: 7,  1.81s/it]
 96%|█████████▌| 14092/14725 [7:09:24<19:07,  1.81s/it]
 96%|█████████▌| 14093/14725 [7:09:25<19:04,  1.81s/it]
 96%|█████████▌| 14094/14725 [7:09:27<19:02,  1.81s/it]
 96%|█████████▌| 14095/14725 [7:09:29<18:56,  1.80s/it]
 96%|█████████▌| 14096/14725 [7:09:31<18:59,  1.81s/it]
 96%|█████████▌| 14097/14725 [7:09:33<18:48,  1.80s/it]
 96%|█████████▌| 14098/14725 [7:09:34<18:54,  1.81s/it]
 96%|█████████▌| 14099/14725 [7:09:36<18:48,  1.80s/it]
 96%|█████████▌| 14100/14725 [7:09:38<18:39,  1.79s/it]
                                                       

 96%|█████████▌| 14100/14725 [7:09:38<18:39,  1.79s/it]
 96%|█████████▌| 14101/14725 [7:09:40<18:38,  1.79s/it]
 96%|█████████▌| 14102/14725 [7:09:42<18:34,  1.79s/it]
 96%|█████████▌| 1410
+0: {'loss': 0.8065, 'grad_norm': 0.6894357356469141, 'learning_rate': 2e-05, 'epoch': 0.96}
+0: 3/14725 [7:09:43<18:37,  1.80s/it]
 96%|█████████▌| 14104/14725 [7:09:45<18:36,  1.80s/it]
 96%|█████████▌| 14105/14725 [7:09:47<18:39,  1.81s/it]
 96%|█████████▌| 14106/14725 [7:09:49<18:27,  1.79s/it]
 96%|█████████▌| 14107/14725 [7:09:51<18:25,  1.79s/it]
 96%|█████████▌| 14108/14725 [7:09:52<18:27,  1.79s/it]
 96%|█████████▌| 14109/14725 [7:09:54<18:19,  1.79s/it]
 96%|█████████▌| 14110/14725 [7:09:56<18:32,  1.81s/it]
                                                       

 96%|█████████▌| 14110/14725 [7:09:56<18:32,  1.81s/it]
 96%|█████████▌| 14111/14725 [7:09:58<18:34,  1.82s/it]
 96%|█████████▌| 14112/14725 [7:10:00<18:30,  1.81s/it]
 96%|█████████▌| 14113/14725 [7:10:01<18:33,  1.82s/it]
 96%|█████████▌| 14114/14725 [7:10:03<18:28,  1.81s/it]
 96%|█████
+0: {'loss': 0.8005, 'grad_norm': 0.7341640833001619, 'learning_rate': 2e-05, 'epoch': 0.96}
+0: ████▌| 14115/14725 [7:10:05<18:37,  1.83s/it]
 96%|█████████▌| 14116/14725 [7:10:07<18:34,  1.83s/it]
 96%|█████████▌| 14117/14725 [7:10:09<18:46,  1.85s/it]
 96%|█████████▌| 14118/14725 [7:10:11<18:39,  1.84s/it]
 96%|█████████▌| 14119/14725 [7:10:13<18:39,  1.85s/it]
 96%|█████████▌| 14120/14725 [7:10:14<18:38,  1.85s/it]
                                                       

 96%|█████████▌| 14120/14725 [7:10:14<18:38,  1.85s/it]
 96%|█████████▌| 14121/14725 [7:10:16<18:37,  1.85s/it]
 96%|█████████▌| 14122/14725 [7:10:18<18:24,  1.83s/it]
 96%|█████████▌| 14123/14725 [7:10:20<18:16,  1.82s/it]
 96%|█████████▌| 14124/14725 [7:10:22<18:02,  1.80s/it]
 96%|█████████▌| 14125/14725 [7:10:23<18:01,  1.80s/it]
 96%|█████████▌| 14126/14725 [7:10:25<18:02,  1.81s/it]
+0: {'loss': 0.8013, 'grad_norm': 0.6790416930648369, 'learning_rate': 2e-05, 'epoch': 0.96}
+0: 
 96%|█████████▌| 14127/14725 [7:10:27<18:01,  1.81s/it]
 96%|█████████▌| 14128/14725 [7:10:29<18:01,  1.81s/it]
 96%|█████████▌| 14129/14725 [7:10:31<18:01,  1.81s/it]
 96%|█████████▌| 14130/14725 [7:10:33<18:07,  1.83s/it]
                                                       

 96%|█████████▌| 14130/14725 [7:10:33<18:07,  1.83s/it]
 96%|█████████▌| 14131/14725 [7:10:34<18:15,  1.84s/it]
 96%|█████████▌| 14132/14725 [7:10:36<18:22,  1.86s/it]
 96%|█████████▌| 14133/14725 [7:10:38<18:23,  1.86s/it]
 96%|█████████▌| 14134/14725 [7:10:40<18:08,  1.84s/it]
 96%|█████████▌| 14135/14725 [7:10:42<17:58,  1.83s/it]
 96%|█████████▌| 14136/14725 [7:10:44<18:01,  1.84s/it]
 96%|█████████▌| 14137/14725 [7:10:45<17:46,  1.81s/it]
 96%|█████████▌| 14138/14725 [7:10
+0: {'loss': 0.8003, 'grad_norm': 0.6599008932588984, 'learning_rate': 2e-05, 'epoch': 0.96}
+0: :47<17:41,  1.81s/it]
 96%|█████████▌| 14139/14725 [7:10:49<17:32,  1.80s/it]
 96%|█████████▌| 14140/14725 [7:10:51<17:47,  1.82s/it]
                                                       

 96%|█████████▌| 14140/14725 [7:10:51<17:47,  1.82s/it]
 96%|█████████▌| 14141/14725 [7:10:53<17:51,  1.83s/it]
 96%|█████████▌| 14142/14725 [7:10:55<17:51,  1.84s/it]
 96%|█████████▌| 14143/14725 [7:10:56<17:35,  1.81s/it]
 96%|█████████▌| 14144/14725 [7:10:58<17:25,  1.80s/it]
 96%|█████████▌| 14145/14725 [7:11:00<17:36,  1.82s/it]
 96%|█████████▌| 14146/14725 [7:11:02<17:36,  1.82s/it]
 96%|█████████▌| 14147/14725 [7:11:04<17:31,  1.82s/it]
 96%|█████████▌| 14148/14725 [7:11:05<17:24,  1.81s/it]
 96%|█████████▌| 14149/14725 [7:11:07<17:24,  1.81s/it]
 96%|█████████�
+0: {'loss': 0.7903, 'grad_norm': 0.6699355412721225, 'learning_rate': 2e-05, 'epoch': 0.96}
+0: {'loss': 0.7924, 'grad_norm': 0.6978387904035028, 'learning_rate': 2e-05, 'epoch': 0.96}
+0: ��| 14150/14725 [7:11:09<17:20,  1.81s/it]
                                                       

 96%|█████████▌| 14150/14725 [7:11:09<17:20,  1.81s/it]
 96%|█████████▌| 14151/14725 [7:11:11<17:16,  1.80s/it]
 96%|█████████▌| 14152/14725 [7:11:13<17:21,  1.82s/it]
 96%|█████████▌| 14153/14725 [7:11:14<17:12,  1.81s/it]
 96%|█████████▌| 14154/14725 [7:11:16<17:10,  1.81s/it]
 96%|█████████▌| 14155/14725 [7:11:18<17:20,  1.83s/it]
 96%|█████████▌| 14156/14725 [7:11:20<17:16,  1.82s/it]
 96%|█████████▌| 14157/14725 [7:11:22<17:09,  1.81s/it]
 96%|█████████▌| 14158/14725 [7:11:24<17:12,  1.82s/it]
 96%|█████████▌| 14159/14725 [7:11:25<16:57,  1.80s/it]
 96%|█████████▌| 14160/14725 [7:11:27<16:52,  1.79s/it]
                                                       

 96%|████████�
+0: {'loss': 0.794, 'grad_norm': 0.6349870000334786, 'learning_rate': 2e-05, 'epoch': 0.96}
+0: �▌| 14160/14725 [7:11:27<16:52,  1.79s/it]
 96%|█████████▌| 14161/14725 [7:11:29<16:47,  1.79s/it]
 96%|█████████▌| 14162/14725 [7:11:31<16:43,  1.78s/it]
 96%|█████████▌| 14163/14725 [7:11:32<16:51,  1.80s/it]
 96%|█████████▌| 14164/14725 [7:11:34<16:53,  1.81s/it]
 96%|█████████▌| 14165/14725 [7:11:36<16:57,  1.82s/it]
 96%|█████████▌| 14166/14725 [7:11:38<17:04,  1.83s/it]
 96%|█████████▌| 14167/14725 [7:11:40<17:02,  1.83s/it]
 96%|█████████▌| 14168/14725 [7:11:42<16:58,  1.83s/it]
 96%|█████████▌| 14169/14725 [7:11:43<16:47,  1.81s/it]
 96%|█████████▌| 14170/14725 [7:11:45<16:59,  1.84s/it]
                                                       

 96%|█████████▌| 14170/14725 [7:11:45<16:59,  1.84s/it]
 96%|█████████▌| 14171/14725 [7:11:47<16:51,  1.83s/it]
 96%|█�
+0: {'loss': 0.8002, 'grad_norm': 0.7110055305900315, 'learning_rate': 2e-05, 'epoch': 0.96}
+0: �███████▌| 14172/14725 [7:11:49<16:56,  1.84s/it]
 96%|█████████▋| 14173/14725 [7:11:51<16:49,  1.83s/it]
 96%|█████████▋| 14174/14725 [7:11:53<16:45,  1.83s/it]
 96%|█████████▋| 14175/14725 [7:11:54<16:33,  1.81s/it]
 96%|█████████▋| 14176/14725 [7:11:56<16:23,  1.79s/it]
 96%|█████████▋| 14177/14725 [7:11:58<16:14,  1.78s/it]
 96%|█████████▋| 14178/14725 [7:12:00<16:12,  1.78s/it]
 96%|█████████▋| 14179/14725 [7:12:01<16:12,  1.78s/it]
 96%|█████████▋| 14180/14725 [7:12:03<16:02,  1.77s/it]
                                                       

 96%|█████████▋| 14180/14725 [7:12:03<16:02,  1.77s/it]
 96%|█████████▋| 14181/14725 [7:12:05<16:11,  1.79s/it]
 96%|█████████▋| 14182/14725 [7:12:07<16:12,  1.79s/it]
 96%|█████████▋| 14183/14725 [7:12:09<16:11, 
+0: {'loss': 0.7815, 'grad_norm': 0.6585545930139768, 'learning_rate': 2e-05, 'epoch': 0.96}
+0:  1.79s/it]
 96%|█████████▋| 14184/14725 [7:12:10<16:05,  1.78s/it]
 96%|█████████▋| 14185/14725 [7:12:12<16:20,  1.82s/it]
 96%|█████████▋| 14186/14725 [7:12:14<16:15,  1.81s/it]
 96%|█████████▋| 14187/14725 [7:12:16<16:14,  1.81s/it]
 96%|█████████▋| 14188/14725 [7:12:18<16:09,  1.81s/it]
 96%|█████████▋| 14189/14725 [7:12:19<16:02,  1.79s/it]
 96%|█████████▋| 14190/14725 [7:12:21<15:56,  1.79s/it]
                                                       

 96%|█████████▋| 14190/14725 [7:12:21<15:56,  1.79s/it]
 96%|█████████▋| 14191/14725 [7:12:23<16:08,  1.81s/it]
 96%|█████████▋| 14192/14725 [7:12:25<15:55,  1.79s/it]
 96%|█████████▋| 14193/14725 [7:12:27<15:59,  1.80s/it]
 96%|█████████▋| 14194/14725 [7:12:28<16:01,  1.81s/it]
 96%|█████████▋| 14195/1
+0: {'loss': 0.7818, 'grad_norm': 0.6697328780565782, 'learning_rate': 2e-05, 'epoch': 0.96}
+0: 4725 [7:12:30<16:03,  1.82s/it]
 96%|█████████▋| 14196/14725 [7:12:32<16:14,  1.84s/it]
 96%|█████████▋| 14197/14725 [7:12:34<16:08,  1.83s/it]
 96%|█████████▋| 14198/14725 [7:12:36<16:09,  1.84s/it]
 96%|█████████▋| 14199/14725 [7:12:38<15:56,  1.82s/it]
 96%|█████████▋| 14200/14725 [7:12:39<15:47,  1.81s/it]
                                                       

 96%|█████████▋| 14200/14725 [7:12:39<15:47,  1.81s/it]
 96%|█████████▋| 14201/14725 [7:12:41<15:52,  1.82s/it]
 96%|█████████▋| 14202/14725 [7:12:43<15:44,  1.81s/it]
 96%|█████████▋| 14203/14725 [7:12:45<15:51,  1.82s/it]
 96%|█████████▋| 14204/14725 [7:12:47<15:43,  1.81s/it]
 96%|█████████▋| 14205/14725 [7:12:48<15:44,  1.82s/it]
 96%|█████████▋| 14206/14725 [7:12:50<15:36,  1.80s/it]
 96%|██████
+0: {'loss': 0.7867, 'grad_norm': 0.6915147144107676, 'learning_rate': 2e-05, 'epoch': 0.97}
+0: ███▋| 14207/14725 [7:12:52<15:44,  1.82s/it]
 96%|█████████▋| 14208/14725 [7:12:54<15:44,  1.83s/it]
 96%|█████████▋| 14209/14725 [7:12:56<15:38,  1.82s/it]
 97%|█████████▋| 14210/14725 [7:12:58<15:35,  1.82s/it]
                                                       

 97%|█████████▋| 14210/14725 [7:12:58<15:35,  1.82s/it]
 97%|█████████▋| 14211/14725 [7:12:59<15:45,  1.84s/it]
 97%|█████████▋| 14212/14725 [7:13:01<15:45,  1.84s/it]
 97%|█████████▋| 14213/14725 [7:13:03<15:39,  1.83s/it]
 97%|█████████▋| 14214/14725 [7:13:05<15:33,  1.83s/it]
 97%|█████████▋| 14215/14725 [7:13:07<15:28,  1.82s/it]
 97%|█████████▋| 14216/14725 [7:13:08<15:14,  1.80s/it]
 97%|█████████▋| 14217/14725 [7:13:10<15:20,  1.81s/it]
 97%|█████████▋| 14218/14725 [7:13:12<15:06,  1.79s/it]
 9
+0: {'loss': 0.7867, 'grad_norm': 0.6469209506609325, 'learning_rate': 2e-05, 'epoch': 0.97}
+0: 7%|█████████▋| 14219/14725 [7:13:14<15:07,  1.79s/it]
 97%|█████████▋| 14220/14725 [7:13:16<15:06,  1.80s/it]
                                                       

 97%|█████████▋| 14220/14725 [7:13:16<15:06,  1.80s/it]
 97%|█████████▋| 14221/14725 [7:13:17<15:04,  1.79s/it]
 97%|█████████▋| 14222/14725 [7:13:19<15:09,  1.81s/it]
 97%|█████████▋| 14223/14725 [7:13:21<15:11,  1.82s/it]
 97%|█████████▋| 14224/14725 [7:13:23<15:09,  1.82s/it]
 97%|█████████▋| 14225/14725 [7:13:25<15:11,  1.82s/it]
 97%|█████████▋| 14226/14725 [7:13:27<15:02,  1.81s/it]
 97%|█████████▋| 14227/14725 [7:13:28<14:58,  1.80s/it]
 97%|█████████▋| 14228/14725 [7:13:30<15:04,  1.82s/it]
 97%|█████████▋| 14229/14725 [7:13:32<15:18,  1.85s/it]
 97%|█████████▋| 14230/14725 [7:13:34
+0: {'loss': 0.7927, 'grad_norm': 0.7037781713524975, 'learning_rate': 2e-05, 'epoch': 0.97}
+0: {'loss': 0.8094, 'grad_norm': 0.6548442581143481, 'learning_rate': 2e-05, 'epoch': 0.97}
+0: <15:16,  1.85s/it]
                                                       

 97%|█████████▋| 14230/14725 [7:13:34<15:16,  1.85s/it]
 97%|█████████▋| 14231/14725 [7:13:36<15:02,  1.83s/it]
 97%|█████████▋| 14232/14725 [7:13:38<14:52,  1.81s/it]
 97%|█████████▋| 14233/14725 [7:13:39<14:47,  1.80s/it]
 97%|█████████▋| 14234/14725 [7:13:41<14:39,  1.79s/it]
 97%|█████████▋| 14235/14725 [7:13:43<14:39,  1.79s/it]
 97%|█████████▋| 14236/14725 [7:13:45<14:49,  1.82s/it]
 97%|█████████▋| 14237/14725 [7:13:47<14:58,  1.84s/it]
 97%|█████████▋| 14238/14725 [7:13:48<14:47,  1.82s/it]
 97%|█████████▋| 14239/14725 [7:13:50<14:59,  1.85s/it]
 97%|█████████▋| 14240/14725 [7:13:52<14:47,  1.83s/it]
                                                       

 97%|█████████▋| 14240/14725 [7:13:
+0: {'loss': 0.7915, 'grad_norm': 0.6386185895186843, 'learning_rate': 2e-05, 'epoch': 0.97}
+0: 52<14:47,  1.83s/it]
 97%|█████████▋| 14241/14725 [7:13:54<14:41,  1.82s/it]
 97%|█████████▋| 14242/14725 [7:13:56<14:33,  1.81s/it]
 97%|█████████▋| 14243/14725 [7:13:58<14:39,  1.82s/it]
 97%|█████████▋| 14244/14725 [7:13:59<14:29,  1.81s/it]
 97%|█████████▋| 14245/14725 [7:14:01<14:37,  1.83s/it]
 97%|█████████▋| 14246/14725 [7:14:03<14:33,  1.82s/it]
 97%|█████████▋| 14247/14725 [7:14:05<14:48,  1.86s/it]
 97%|█████████▋| 14248/14725 [7:14:07<14:38,  1.84s/it]
 97%|█████████▋| 14249/14725 [7:14:09<14:37,  1.84s/it]
 97%|█████████▋| 14250/14725 [7:14:10<14:26,  1.82s/it]
                                                       

 97%|█████████▋| 14250/14725 [7:14:10<14:26,  1.82s/it]
 97%|█████████▋| 14251/14725 [7:14:12<14:17,  1.81s/it]
 97%|█████████�
+0: {'loss': 0.7917, 'grad_norm': 0.6609935807876003, 'learning_rate': 2e-05, 'epoch': 0.97}
+0: �| 14252/14725 [7:14:14<14:16,  1.81s/it]
 97%|█████████▋| 14253/14725 [7:14:16<14:09,  1.80s/it]
 97%|█████████▋| 14254/14725 [7:14:18<14:14,  1.81s/it]
 97%|█████████▋| 14255/14725 [7:14:19<14:02,  1.79s/it]
 97%|█████████▋| 14256/14725 [7:14:21<14:04,  1.80s/it]
 97%|█████████▋| 14257/14725 [7:14:23<13:58,  1.79s/it]
 97%|█████████▋| 14258/14725 [7:14:25<13:55,  1.79s/it]
 97%|█████████▋| 14259/14725 [7:14:27<14:02,  1.81s/it]
 97%|█████████▋| 14260/14725 [7:14:28<13:55,  1.80s/it]
                                                       

 97%|█████████▋| 14260/14725 [7:14:28<13:55,  1.80s/it]
 97%|█████████▋| 14261/14725 [7:14:30<13:56,  1.80s/it]
 97%|█████████▋| 14262/14725 [7:14:32<14:02,  1.82s/it]
 97%|█████████▋| 14263/14725 [7:14:34<13:49,  1.80s/it]
 97%|██�
+0: {'loss': 0.7953, 'grad_norm': 0.7143950641271, 'learning_rate': 2e-05, 'epoch': 0.97}
+0: �██████▋| 14264/14725 [7:14:36<13:41,  1.78s/it]
 97%|█████████▋| 14265/14725 [7:14:37<13:38,  1.78s/it]
 97%|█████████▋| 14266/14725 [7:14:39<13:33,  1.77s/it]
 97%|█████████▋| 14267/14725 [7:14:41<13:31,  1.77s/it]
 97%|█████████▋| 14268/14725 [7:14:43<13:28,  1.77s/it]
 97%|█████████▋| 14269/14725 [7:14:44<13:22,  1.76s/it]
 97%|█████████▋| 14270/14725 [7:14:46<13:21,  1.76s/it]
                                                       

 97%|█████████▋| 14270/14725 [7:14:46<13:21,  1.76s/it]
 97%|█████████▋| 14271/14725 [7:14:48<13:29,  1.78s/it]
 97%|█████████▋| 14272/14725 [7:14:50<13:28,  1.78s/it]
 97%|█████████▋| 14273/14725 [7:14:51<13:22,  1.78s/it]
 97%|█████████▋| 14274/14725 [7:14:53<13:21,  1.78s/it]
 97%|█████████▋| 14275/14725 [7:14:55<13:18,  1.
+0: {'loss': 0.7928, 'grad_norm': 0.6913621425587284, 'learning_rate': 2e-05, 'epoch': 0.97}
+0: 77s/it]
 97%|█████████▋| 14276/14725 [7:14:57<13:17,  1.78s/it]
 97%|█████████▋| 14277/14725 [7:14:59<13:17,  1.78s/it]
 97%|█████████▋| 14278/14725 [7:15:00<13:10,  1.77s/it]
 97%|█████████▋| 14279/14725 [7:15:02<13:23,  1.80s/it]
 97%|█████████▋| 14280/14725 [7:15:04<13:24,  1.81s/it]
                                                       

 97%|█████████▋| 14280/14725 [7:15:04<13:24,  1.81s/it]
 97%|█████████▋| 14281/14725 [7:15:06<13:30,  1.83s/it]
 97%|█████████▋| 14282/14725 [7:15:08<13:26,  1.82s/it]
 97%|█████████▋| 14283/14725 [7:15:09<13:20,  1.81s/it]
 97%|█████████▋| 14284/14725 [7:15:11<13:25,  1.83s/it]
 97%|█████████▋| 14285/14725 [7:15:13<13:16,  1.81s/it]
 97%|█████████▋| 14286/14725 [7:15:15<13:16,  1.81s/it]
 97%|█████████▋| 14287/1472
+0: {'loss': 0.7887, 'grad_norm': 0.6581640869127526, 'learning_rate': 2e-05, 'epoch': 0.97}
+0: 5 [7:15:17<13:10,  1.80s/it]
 97%|█████████▋| 14288/14725 [7:15:19<13:11,  1.81s/it]
 97%|█████████▋| 14289/14725 [7:15:20<13:03,  1.80s/it]
 97%|█████████▋| 14290/14725 [7:15:22<13:06,  1.81s/it]
                                                       

 97%|█████████▋| 14290/14725 [7:15:22<13:06,  1.81s/it]
 97%|█████████▋| 14291/14725 [7:15:24<13:10,  1.82s/it]
 97%|█████████▋| 14292/14725 [7:15:26<13:10,  1.83s/it]
 97%|█████████▋| 14293/14725 [7:15:28<13:00,  1.81s/it]
 97%|█████████▋| 14294/14725 [7:15:29<13:01,  1.81s/it]
 97%|█████████▋| 14295/14725 [7:15:31<12:56,  1.80s/it]
 97%|█████████▋| 14296/14725 [7:15:33<13:08,  1.84s/it]
 97%|█████████▋| 14297/14725 [7:15:35<13:05,  1.84s/it]
 97%|█████████▋| 14298/14725 [7:15:37<13:01,  1.83s/it]
 97%|███████
+0: {'loss': 0.7885, 'grad_norm': 0.6905660204655699, 'learning_rate': 2e-05, 'epoch': 0.97}
+0: ██▋| 14299/14725 [7:15:39<13:08,  1.85s/it]
 97%|█████████▋| 14300/14725 [7:15:40<12:57,  1.83s/it]
                                                       

 97%|█████████▋| 14300/14725 [7:15:40<12:57,  1.83s/it]
 97%|█████████▋| 14301/14725 [7:15:42<12:52,  1.82s/it]
 97%|█████████▋| 14302/14725 [7:15:44<12:43,  1.80s/it]
 97%|█████████▋| 14303/14725 [7:15:46<12:38,  1.80s/it]
 97%|█████████▋| 14304/14725 [7:15:48<12:35,  1.79s/it]
 97%|█████████▋| 14305/14725 [7:15:49<12:32,  1.79s/it]
 97%|█████████▋| 14306/14725 [7:15:51<12:26,  1.78s/it]
 97%|█████████▋| 14307/14725 [7:15:53<12:36,  1.81s/it]
 97%|█████████▋| 14308/14725 [7:15:55<12:31,  1.80s/it]
 97%|█████████▋| 14309/14725 [7:15:57<12:22,  1.78s/it]
 97%|█████████▋| 14310/14725 [7:15:58<12:15,  1.77s/it]
     
+0: {'loss': 0.7861, 'grad_norm': 0.6689947835574724, 'learning_rate': 2e-05, 'epoch': 0.97}
+0: {'loss': 0.7879, 'grad_norm': 0.6558539454291415, 'learning_rate': 2e-05, 'epoch': 0.97}
+0:                                                   

 97%|█████████▋| 14310/14725 [7:15:58<12:15,  1.77s/it]
 97%|█████████▋| 14311/14725 [7:16:00<12:15,  1.78s/it]
 97%|█████████▋| 14312/14725 [7:16:02<12:09,  1.77s/it]
 97%|█████████▋| 14313/14725 [7:16:04<12:03,  1.76s/it]
 97%|█████████▋| 14314/14725 [7:16:05<12:01,  1.76s/it]
 97%|█████████▋| 14315/14725 [7:16:07<12:09,  1.78s/it]
 97%|█████████▋| 14316/14725 [7:16:09<12:16,  1.80s/it]
 97%|█████████▋| 14317/14725 [7:16:11<12:17,  1.81s/it]
 97%|█████████▋| 14318/14725 [7:16:13<12:17,  1.81s/it]
 97%|█████████▋| 14319/14725 [7:16:14<12:18,  1.82s/it]
 97%|█████████▋| 14320/14725 [7:16:16<12:07,  1.80s/it]
                                                       

 97%|█████████▋| 14320/14725 [7:16:16<12:07,  1.80s/it]
 97
+0: {'loss': 0.7726, 'grad_norm': 0.6429417661162213, 'learning_rate': 2e-05, 'epoch': 0.97}
+0: %|█████████▋| 14321/14725 [7:16:18<12:08,  1.80s/it]
 97%|█████████▋| 14322/14725 [7:16:20<11:59,  1.79s/it]
 97%|█████████▋| 14323/14725 [7:16:22<12:01,  1.80s/it]
 97%|█████████▋| 14324/14725 [7:16:23<11:58,  1.79s/it]
 97%|█████████▋| 14325/14725 [7:16:25<11:53,  1.78s/it]
 97%|█████████▋| 14326/14725 [7:16:27<11:48,  1.77s/it]
 97%|█████████▋| 14327/14725 [7:16:29<12:01,  1.81s/it]
 97%|█████████▋| 14328/14725 [7:16:31<11:56,  1.80s/it]
 97%|█████████▋| 14329/14725 [7:16:32<11:57,  1.81s/it]
 97%|█████████▋| 14330/14725 [7:16:34<11:50,  1.80s/it]
                                                       

 97%|█████████▋| 14330/14725 [7:16:34<11:50,  1.80s/it]
 97%|█████████▋| 14331/14725 [7:16:36<11:47,  1.80s/it]
 97%|█████████▋| 14332/14725 [7:16:38<
+0: {'loss': 0.81, 'grad_norm': 0.6636719059215107, 'learning_rate': 2e-05, 'epoch': 0.97}
+0: 11:40,  1.78s/it]
 97%|█████████▋| 14333/14725 [7:16:40<11:45,  1.80s/it]
 97%|█████████▋| 14334/14725 [7:16:41<11:51,  1.82s/it]
 97%|█████████▋| 14335/14725 [7:16:43<11:50,  1.82s/it]
 97%|█████████▋| 14336/14725 [7:16:45<11:43,  1.81s/it]
 97%|█████████▋| 14337/14725 [7:16:47<11:40,  1.81s/it]
 97%|█████████▋| 14338/14725 [7:16:49<11:39,  1.81s/it]
 97%|█████████▋| 14339/14725 [7:16:50<11:36,  1.80s/it]
 97%|█████████▋| 14340/14725 [7:16:52<11:33,  1.80s/it]
                                                       

 97%|█████████▋| 14340/14725 [7:16:52<11:33,  1.80s/it]
 97%|█████████▋| 14341/14725 [7:16:54<11:31,  1.80s/it]
 97%|█████████▋| 14342/14725 [7:16:56<11:33,  1.81s/it]
 97%|█████████▋| 14343/14725 [7:16:58<11:32,  1.81s/it]
 97%|█████████▋| 
+0: {'loss': 0.7864, 'grad_norm': 0.6773038383825635, 'learning_rate': 2e-05, 'epoch': 0.97}
+0: 14344/14725 [7:17:00<11:32,  1.82s/it]
 97%|█████████▋| 14345/14725 [7:17:01<11:21,  1.79s/it]
 97%|█████████▋| 14346/14725 [7:17:03<11:25,  1.81s/it]
 97%|█████████▋| 14347/14725 [7:17:05<11:20,  1.80s/it]
 97%|█████████▋| 14348/14725 [7:17:07<11:25,  1.82s/it]
 97%|█████████▋| 14349/14725 [7:17:09<11:18,  1.80s/it]
 97%|█████████▋| 14350/14725 [7:17:10<11:16,  1.80s/it]
                                                       

 97%|█████████▋| 14350/14725 [7:17:10<11:16,  1.80s/it]
 97%|█████████▋| 14351/14725 [7:17:12<11:12,  1.80s/it]
 97%|█████████▋| 14352/14725 [7:17:14<11:13,  1.80s/it]
 97%|█████████▋| 14353/14725 [7:17:16<11:08,  1.80s/it]
 97%|█████████▋| 14354/14725 [7:17:18<11:25,  1.85s/it]
 97%|█████████▋| 14355/14725 [7:17:20<11:25,  1.85s/it]
 97%|███�
+0: {'loss': 0.7931, 'grad_norm': 0.6908363182992964, 'learning_rate': 2e-05, 'epoch': 0.98}
+0: �█████▋| 14356/14725 [7:17:21<11:22,  1.85s/it]
 98%|█████████▊| 14357/14725 [7:17:23<11:10,  1.82s/it]
 98%|█████████▊| 14358/14725 [7:17:25<11:00,  1.80s/it]
 98%|█████████▊| 14359/14725 [7:17:27<10:54,  1.79s/it]
 98%|█████████▊| 14360/14725 [7:17:28<10:50,  1.78s/it]
                                                       

 98%|█████████▊| 14360/14725 [7:17:28<10:50,  1.78s/it]
 98%|█████████▊| 14361/14725 [7:17:30<10:55,  1.80s/it]
 98%|█████████▊| 14362/14725 [7:17:32<10:48,  1.79s/it]
 98%|█████████▊| 14363/14725 [7:17:34<10:48,  1.79s/it]
 98%|█████████▊| 14364/14725 [7:17:36<10:45,  1.79s/it]
 98%|█████████▊| 14365/14725 [7:17:37<10:40,  1.78s/it]
 98%|█████████▊| 14366/14725 [7:17:39<10:41,  1.79s/it]
 98%|█████████▊| 14367/14725 [7:17:41<10:34,  1.77s
+0: {'loss': 0.796, 'grad_norm': 0.7156812660667937, 'learning_rate': 2e-05, 'epoch': 0.98}
+0: /it]
 98%|█████████▊| 14368/14725 [7:17:43<10:33,  1.77s/it]
 98%|█████████▊| 14369/14725 [7:17:44<10:27,  1.76s/it]
 98%|█████████▊| 14370/14725 [7:17:46<10:31,  1.78s/it]
                                                       

 98%|█████████▊| 14370/14725 [7:17:46<10:31,  1.78s/it]
 98%|█████████▊| 14371/14725 [7:17:48<10:29,  1.78s/it]
 98%|█████████▊| 14372/14725 [7:17:50<10:33,  1.80s/it]
 98%|█████████▊| 14373/14725 [7:17:52<10:33,  1.80s/it]
 98%|█████████▊| 14374/14725 [7:17:53<10:28,  1.79s/it]
 98%|█████████▊| 14375/14725 [7:17:55<10:23,  1.78s/it]
 98%|█████████▊| 14376/14725 [7:17:57<10:24,  1.79s/it]
 98%|█████████▊| 14377/14725 [7:17:59<10:33,  1.82s/it]
 98%|█████████▊| 14378/14725 [7:18:01<10:25,  1.80s/it]
 98%|█████████▊| 14379/14725 [
+0: {'loss': 0.7958, 'grad_norm': 0.6775408499568508, 'learning_rate': 2e-05, 'epoch': 0.98}
+0: 7:18:02<10:22,  1.80s/it]
 98%|█████████▊| 14380/14725 [7:18:04<10:29,  1.82s/it]
                                                       

 98%|█████████▊| 14380/14725 [7:18:04<10:29,  1.82s/it]
 98%|█████████▊| 14381/14725 [7:18:06<10:23,  1.81s/it]
 98%|█████████▊| 14382/14725 [7:18:08<10:15,  1.80s/it]
 98%|█████████▊| 14383/14725 [7:18:10<10:09,  1.78s/it]
 98%|█████████▊| 14384/14725 [7:18:11<10:06,  1.78s/it]
 98%|█████████▊| 14385/14725 [7:18:13<10:06,  1.78s/it]
 98%|█████████▊| 14386/14725 [7:18:15<10:00,  1.77s/it]
 98%|█████████▊| 14387/14725 [7:18:17<10:07,  1.80s/it]
 98%|█████████▊| 14388/14725 [7:18:19<10:14,  1.82s/it]
 98%|█████████▊| 14389/14725 [7:18:20<10:07,  1.81s/it]
 98%|█████████▊| 14390/14725 [7:18:22<10:06,  1.81s/it]
                             
+0: {'loss': 0.7969, 'grad_norm': 0.6864477368967102, 'learning_rate': 2e-05, 'epoch': 0.98}
+0: {'loss': 0.7764, 'grad_norm': 0.6756363101449374, 'learning_rate': 2e-05, 'epoch': 0.98}
+0:                           

 98%|█████████▊| 14390/14725 [7:18:22<10:06,  1.81s/it]
 98%|█████████▊| 14391/14725 [7:18:24<10:14,  1.84s/it]
 98%|█████████▊| 14392/14725 [7:18:26<10:05,  1.82s/it]
 98%|█████████▊| 14393/14725 [7:18:28<10:01,  1.81s/it]
 98%|█████████▊| 14394/14725 [7:18:30<09:58,  1.81s/it]
 98%|█████████▊| 14395/14725 [7:18:31<09:51,  1.79s/it]
 98%|█████████▊| 14396/14725 [7:18:33<09:53,  1.80s/it]
 98%|█████████▊| 14397/14725 [7:18:35<09:48,  1.79s/it]
 98%|█████████▊| 14398/14725 [7:18:37<09:54,  1.82s/it]
 98%|█████████▊| 14399/14725 [7:18:39<09:50,  1.81s/it]
 98%|█████████▊| 14400/14725 [7:18:40<09:44,  1.80s/it]
                                                       

 98%|█████████▊| 14400/14725 [7:18:40<09:44,  1.80s/it]
 98%|███████�
+0: {'loss': 0.7774, 'grad_norm': 0.6453317713522245, 'learning_rate': 2e-05, 'epoch': 0.98}
+0: ��█▊| 14401/14725 [7:18:42<09:51,  1.82s/it]
 98%|█████████▊| 14402/14725 [7:18:44<09:47,  1.82s/it]
 98%|█████████▊| 14403/14725 [7:18:46<09:40,  1.80s/it]
 98%|█████████▊| 14404/14725 [7:18:48<09:35,  1.79s/it]
 98%|█████████▊| 14405/14725 [7:18:49<09:30,  1.78s/it]
 98%|█████████▊| 14406/14725 [7:18:51<09:33,  1.80s/it]
 98%|█████████▊| 14407/14725 [7:18:53<09:30,  1.79s/it]
 98%|█████████▊| 14408/14725 [7:18:55<09:29,  1.80s/it]
 98%|█████████▊| 14409/14725 [7:18:57<09:36,  1.82s/it]
 98%|█████████▊| 14410/14725 [7:18:58<09:35,  1.83s/it]
                                                       

 98%|█████████▊| 14410/14725 [7:18:58<09:35,  1.83s/it]
 98%|█████████▊| 14411/14725 [7:19:00<09:35,  1.83s/it]
 98%|█████████▊| 14412/14725 [7:19:02<09:30,  1.82s/it]
 98%|�
+0: {'loss': 0.8149, 'grad_norm': 0.6494919912000723, 'learning_rate': 2e-05, 'epoch': 0.98}
+0: ��████████▊| 14413/14725 [7:19:04<09:27,  1.82s/it]
 98%|█████████▊| 14414/14725 [7:19:06<09:30,  1.83s/it]
 98%|█████████▊| 14415/14725 [7:19:08<09:28,  1.83s/it]
 98%|█████████▊| 14416/14725 [7:19:09<09:20,  1.82s/it]
 98%|█████████▊| 14417/14725 [7:19:11<09:18,  1.81s/it]
 98%|█████████▊| 14418/14725 [7:19:13<09:17,  1.82s/it]
 98%|█████████▊| 14419/14725 [7:19:15<09:13,  1.81s/it]
 98%|█████████▊| 14420/14725 [7:19:17<09:10,  1.80s/it]
                                                       

 98%|█████████▊| 14420/14725 [7:19:17<09:10,  1.80s/it]
 98%|█████████▊| 14421/14725 [7:19:18<09:09,  1.81s/it]
 98%|█████████▊| 14422/14725 [7:19:20<09:09,  1.81s/it]
 98%|█████████▊| 14423/14725 [7:19:22<09:06,  1.81s/it]
 98%|█████████▊| 14424/14725 [7:19:24<09:
+0: {'loss': 0.7895, 'grad_norm': 0.6208772882613314, 'learning_rate': 1.9992105470889727e-05, 'epoch': 0.98}
+0: 06,  1.81s/it]
 98%|█████████▊| 14425/14725 [7:19:26<08:57,  1.79s/it]
 98%|█████████▊| 14426/14725 [7:19:27<08:51,  1.78s/it]
 98%|█████████▊| 14427/14725 [7:19:29<08:50,  1.78s/it]
 98%|█████████▊| 14428/14725 [7:19:31<08:53,  1.80s/it]
 98%|█████████▊| 14429/14725 [7:19:33<08:54,  1.81s/it]
 98%|█████████▊| 14430/14725 [7:19:35<09:04,  1.85s/it]
                                                       

 98%|█████████▊| 14430/14725 [7:19:35<09:04,  1.85s/it]
 98%|█████████▊| 14431/14725 [7:19:36<08:56,  1.82s/it]
 98%|█████████▊| 14432/14725 [7:19:38<08:50,  1.81s/it]
 98%|█████████▊| 14433/14725 [7:19:40<08:44,  1.79s/it]
 98%|█████████▊| 14434/14725 [7:19:42<08:52,  1.83s/it]
 98%|█████████▊| 14435/14725 [7:19:44<08:47,  1.82s/it]
 98%|█████████▊| 144
+0: {'loss': 0.7872, 'grad_norm': 0.6281318855959953, 'learning_rate': 1.9903450996666898e-05, 'epoch': 0.98}
+0: 36/14725 [7:19:45<08:41,  1.80s/it]
 98%|█████████▊| 14437/14725 [7:19:47<08:46,  1.83s/it]
 98%|█████████▊| 14438/14725 [7:19:49<08:51,  1.85s/it]
 98%|█████████▊| 14439/14725 [7:19:51<08:44,  1.83s/it]
 98%|█████████▊| 14440/14725 [7:19:53<08:36,  1.81s/it]
                                                       

 98%|█████████▊| 14440/14725 [7:19:53<08:36,  1.81s/it]
 98%|█████████▊| 14441/14725 [7:19:55<08:37,  1.82s/it]
 98%|█████████▊| 14442/14725 [7:19:56<08:34,  1.82s/it]
 98%|█████████▊| 14443/14725 [7:19:58<08:26,  1.80s/it]
 98%|█████████▊| 14444/14725 [7:20:00<08:20,  1.78s/it]
 98%|█████████▊| 14445/14725 [7:20:02<08:24,  1.80s/it]
 98%|█████████▊| 14446/14725 [7:20:04<08:22,  1.80s/it]
 98%|█████████▊| 14447/14725 [7:20:05<08:21,  1.80s/it]
 98%|████�
+0: {'loss': 0.7876, 'grad_norm': 0.6843540391243078, 'learning_rate': 1.9717248450157682e-05, 'epoch': 0.98}
+0: �████▊| 14448/14725 [7:20:07<08:27,  1.83s/it]
 98%|█████████▊| 14449/14725 [7:20:09<08:24,  1.83s/it]
 98%|█████████▊| 14450/14725 [7:20:11<08:21,  1.82s/it]
                                                       

 98%|█████████▊| 14450/14725 [7:20:11<08:21,  1.82s/it]
 98%|█████████▊| 14451/14725 [7:20:13<08:13,  1.80s/it]
 98%|█████████▊| 14452/14725 [7:20:15<08:10,  1.80s/it]
 98%|█████████▊| 14453/14725 [7:20:16<08:11,  1.81s/it]
 98%|█████████▊| 14454/14725 [7:20:18<08:07,  1.80s/it]
 98%|█████████▊| 14455/14725 [7:20:20<08:08,  1.81s/it]
 98%|█████████▊| 14456/14725 [7:20:22<08:05,  1.81s/it]
 98%|█████████▊| 14457/14725 [7:20:24<08:06,  1.81s/it]
 98%|█████████▊| 14458/14725 [7:20:25<08:02,  1.81s/it]
 98%|█████████▊| 14459/14725 [7:20:27<08:00,  1.81s/it
+0: {'loss': 0.7702, 'grad_norm': 0.7191747039639813, 'learning_rate': 1.9435537905427028e-05, 'epoch': 0.98}
+0: ]
 98%|█████████▊| 14460/14725 [7:20:29<07:59,  1.81s/it]
                                                       

 98%|█████████▊| 14460/14725 [7:20:29<07:59,  1.81s/it]
 98%|█████████▊| 14461/14725 [7:20:31<07:55,  1.80s/it]
 98%|█████████▊| 14462/14725 [7:20:33<07:59,  1.82s/it]
 98%|█████████▊| 14463/14725 [7:20:34<07:56,  1.82s/it]
 98%|█████████▊| 14464/14725 [7:20:36<07:48,  1.80s/it]
 98%|█████████▊| 14465/14725 [7:20:38<07:47,  1.80s/it]
 98%|█████████▊| 14466/14725 [7:20:40<07:46,  1.80s/it]
 98%|█████████▊| 14467/14725 [7:20:42<07:51,  1.83s/it]
 98%|█████████▊| 14468/14725 [7:20:43<07:44,  1.81s/it]
 98%|█████████▊| 14469/14725 [7:20:45<07:53,  1.85s/it]
 98%|█████████▊| 14470/14725 [7:20:47<07:44,  1.82s/it]
                                                     
+0: {'loss': 0.7894, 'grad_norm': 0.6445267409074774, 'learning_rate': 1.9061405842154716e-05, 'epoch': 0.98}
+0: {'loss': 0.785, 'grad_norm': 0.6352726316725593, 'learning_rate': 1.859895132951814e-05, 'epoch': 0.98}
+0:   

 98%|█████████▊| 14470/14725 [7:20:47<07:44,  1.82s/it]
 98%|█████████▊| 14471/14725 [7:20:49<07:43,  1.83s/it]
 98%|█████████▊| 14472/14725 [7:20:51<07:37,  1.81s/it]
 98%|█████████▊| 14473/14725 [7:20:53<07:33,  1.80s/it]
 98%|█████████▊| 14474/14725 [7:20:54<07:27,  1.78s/it]
 98%|█████████▊| 14475/14725 [7:20:56<07:24,  1.78s/it]
 98%|█████████▊| 14476/14725 [7:20:58<07:18,  1.76s/it]
 98%|█████████▊| 14477/14725 [7:21:00<07:19,  1.77s/it]
 98%|█████████▊| 14478/14725 [7:21:01<07:20,  1.78s/it]
 98%|█████████▊| 14479/14725 [7:21:03<07:22,  1.80s/it]
 98%|█████████▊| 14480/14725 [7:21:05<07:19,  1.80s/it]
                                                       

 98%|█████████▊| 14480/14725 [7:21:05<07:19,  1.80s/it]
 98%|█████████▊| 14481/14725 [7
+0: {'loss': 0.7731, 'grad_norm': 0.714974941412266, 'learning_rate': 1.805324111593256e-05, 'epoch': 0.98}
+0: :21:07<07:14,  1.78s/it]
 98%|█████████▊| 14482/14725 [7:21:09<07:22,  1.82s/it]
 98%|█████████▊| 14483/14725 [7:21:10<07:17,  1.81s/it]
 98%|█████████▊| 14484/14725 [7:21:12<07:18,  1.82s/it]
 98%|█████████▊| 14485/14725 [7:21:14<07:10,  1.79s/it]
 98%|█████████▊| 14486/14725 [7:21:16<07:08,  1.79s/it]
 98%|█████████▊| 14487/14725 [7:21:18<07:02,  1.78s/it]
 98%|█████████▊| 14488/14725 [7:21:19<06:58,  1.77s/it]
 98%|█████████▊| 14489/14725 [7:21:21<06:53,  1.75s/it]
 98%|█████████▊| 14490/14725 [7:21:23<06:51,  1.75s/it]
                                                       

 98%|█████████▊| 14490/14725 [7:21:23<06:51,  1.75s/it]
 98%|█████████▊| 14491/14725 [7:21:25<06:56,  1.78s/it]
 98%|█████████▊| 14492/14725 [7:21:26<06:52,  1.77s/it]
 98%|████████�
+0: {'loss': 0.7818, 'grad_norm': 0.6421909853875409, 'learning_rate': 1.743025411669523e-05, 'epoch': 0.98}
+0: ��▊| 14493/14725 [7:21:28<06:48,  1.76s/it]
 98%|█████████▊| 14494/14725 [7:21:30<06:46,  1.76s/it]
 98%|█████████▊| 14495/14725 [7:21:32<06:44,  1.76s/it]
 98%|█████████▊| 14496/14725 [7:21:33<06:42,  1.76s/it]
 98%|█████████▊| 14497/14725 [7:21:35<06:39,  1.75s/it]
 98%|█████████▊| 14498/14725 [7:21:37<06:35,  1.74s/it]
 98%|█████████▊| 14499/14725 [7:21:39<06:36,  1.75s/it]
 98%|█████████▊| 14500/14725 [7:21:40<06:34,  1.75s/it]
                                                       

 98%|█████████▊| 14500/14725 [7:21:40<06:34,  1.75s/it]
 98%|█████████▊| 14501/14725 [7:21:42<06:32,  1.75s/it]
 98%|█████████▊| 14502/14725 [7:21:44<06:33,  1.77s/it]
 98%|█████████▊| 14503/14725 [7:21:46<06:30,  1.76s/it]
 98%|█████████▊| 14504/14725 [7:21:47<06:29,  1.76s/it]
 99%|█�
+0: {'loss': 0.7811, 'grad_norm': 0.6646966126730368, 'learning_rate': 1.673681590773821e-05, 'epoch': 0.99}
+0: ��███████▊| 14505/14725 [7:21:49<06:29,  1.77s/it]
 99%|█████████▊| 14506/14725 [7:21:51<06:29,  1.78s/it]
 99%|███████���█▊| 14507/14725 [7:21:53<06:29,  1.79s/it]
 99%|█████████▊| 14508/14725 [7:21:55<06:30,  1.80s/it]
 99%|█████████▊| 14509/14725 [7:21:56<06:27,  1.79s/it]
 99%|█████████▊| 14510/14725 [7:21:58<06:28,  1.80s/it]
                                                       

 99%|█████████▊| 14510/14725 [7:21:58<06:28,  1.80s/it]
 99%|█████████▊| 14511/14725 [7:22:00<06:24,  1.80s/it]
 99%|█████████▊| 14512/14725 [7:22:02<06:22,  1.79s/it]
 99%|█████████▊| 14513/14725 [7:22:04<06:15,  1.77s/it]
 99%|█████████▊| 14514/14725 [7:22:05<06:14,  1.77s/it]
 99%|█████████▊| 14515/14725 [7:22:07<06:22,  1.82s/it]
 99%|█████████▊| 14516/14725 [7:22:09<06:22,
+0: {'loss': 0.7897, 'grad_norm': 0.6080425974340008, 'learning_rate': 1.59805239431901e-05, 'epoch': 0.99}
+0:   1.83s/it]
 99%|█████████▊| 14517/14725 [7:22:11<06:15,  1.80s/it]
 99%|█████████▊| 14518/14725 [7:22:13<06:08,  1.78s/it]
 99%|█████████▊| 14519/14725 [7:22:14<06:08,  1.79s/it]
 99%|█████████▊| 14520/14725 [7:22:16<06:04,  1.78s/it]
                                                       

 99%|█████████▊| 14520/14725 [7:22:16<06:04,  1.78s/it]
 99%|█████████▊| 14521/14725 [7:22:18<06:01,  1.77s/it]
 99%|█████████▊| 14522/14725 [7:22:20<06:02,  1.79s/it]
 99%|█████████▊| 14523/14725 [7:22:22<06:03,  1.80s/it]
 99%|█████████▊| 14524/14725 [7:22:23<06:05,  1.82s/it]
 99%|█████████▊| 14525/14725 [7:22:25<06:01,  1.81s/it]
 99%|█████████▊| 14526/14725 [7:22:27<05:54,  1.78s/it]
 99%|█████████▊| 14527/14725 [7:22:29<05:51,  1.78s/it]
 99%|█████████▊| 14528/
+0: {'loss': 0.7942, 'grad_norm': 0.6797900633741023, 'learning_rate': 1.5169664316078758e-05, 'epoch': 0.99}
+0: 14725 [7:22:30<05:48,  1.77s/it]
 99%|█████████▊| 14529/14725 [7:22:32<05:43,  1.76s/it]
 99%|█████████▊| 14530/14725 [7:22:34<05:51,  1.80s/it]
                                                       

 99%|█████████▊| 14530/14725 [7:22:34<05:51,  1.80s/it]
 99%|█████████▊| 14531/14725 [7:22:36<05:47,  1.79s/it]
 99%|█████████▊| 14532/14725 [7:22:38<05:47,  1.80s/it]
 99%|█████████▊| 14533/14725 [7:22:39<05:41,  1.78s/it]
 99%|█████████▊| 14534/14725 [7:22:41<05:36,  1.76s/it]
 99%|█████████▊| 14535/14725 [7:22:43<05:37,  1.78s/it]
 99%|█████████▊| 14536/14725 [7:22:45<05:32,  1.76s/it]
 99%|█████████▊| 14537/14725 [7:22:46<05:30,  1.76s/it]
 99%|█████████▊| 14538/14725 [7:22:48<05:34,  1.79s/it]
 99%|█████████▊| 14539/14725 [7:22:50<05:29,  1.77s/it]
 99%|█████�
+0: {'loss': 0.7918, 'grad_norm': 0.6357089276411193, 'learning_rate': 1.4313120974162105e-05, 'epoch': 0.99}
+0: {'loss': 0.7832, 'grad_norm': 0.6360917279879135, 'learning_rate': 1.342027838553739e-05, 'epoch': 0.99}
+0: �███▊| 14540/14725 [7:22:52<05:27,  1.77s/it]
                                                       

 99%|█████████▊| 14540/14725 [7:22:52<05:27,  1.77s/it]
 99%|█████████▉| 14541/14725 [7:22:54<05:28,  1.78s/it]
 99%|█████████▉| 14542/14725 [7:22:55<05:24,  1.78s/it]
 99%|█████████▉| 14543/14725 [7:22:57<05:22,  1.77s/it]
 99%|█████████▉| 14544/14725 [7:22:59<05:25,  1.80s/it]
 99%|█████████▉| 14545/14725 [7:23:01<05:21,  1.79s/it]
 99%|█████████▉| 14546/14725 [7:23:03<05:20,  1.79s/it]
 99%|█████████▉| 14547/14725 [7:23:04<05:14,  1.77s/it]
 99%|█████████▉| 14548/14725 [7:23:06<05:12,  1.76s/it]
 99%|█████████▉| 14549/14725 [7:23:08<05:17,  1.80s/it]
 99%|█████████▉| 14550/14725 [7:23:10<05:16,  1.81s/it]
                                                       

 99%|█████
+0: {'loss': 0.7871, 'grad_norm': 0.6103709334955774, 'learning_rate': 1.250091872044492e-05, 'epoch': 0.99}
+0: ████▉| 14550/14725 [7:23:10<05:16,  1.81s/it]
 99%|█████████▉| 14551/14725 [7:23:11<05:11,  1.79s/it]
 99%|█████████▉| 14552/14725 [7:23:13<05:10,  1.79s/it]
 99%|█████████▉| 14553/14725 [7:23:15<05:08,  1.79s/it]
 99%|█████████▉| 14554/14725 [7:23:17<05:10,  1.82s/it]
 99%|█████████▉| 14555/14725 [7:23:19<05:08,  1.81s/it]
 99%|█████████▉| 14556/14725 [7:23:21<05:05,  1.81s/it]
 99%|█████████▉| 14557/14725 [7:23:22<05:02,  1.80s/it]
 99%|█████████▉| 14558/14725 [7:23:24<05:00,  1.80s/it]
 99%|█████████▉| 14559/14725 [7:23:26<05:03,  1.83s/it]
 99%|█████████▉| 14560/14725 [7:23:28<04:58,  1.81s/it]
                                                       

 99%|█████████▉| 14560/14725 [7:23:28<04:58,  1.81s/it]
 99%|█████████▉| 14561/14725 [7:23:29<04:53,  1.79s/it]
+0: {'loss': 0.7838, 'grad_norm': 0.6033173806131285, 'learning_rate': 1.1565114675763823e-05, 'epoch': 0.99}
+0: 
 99%|█████████▉| 14562/14725 [7:23:31<04:51,  1.79s/it]
 99%|█████████▉| 14563/14725 [7:23:33<04:52,  1.80s/it]
 99%|█████████▉| 14564/14725 [7:23:35<04:52,  1.82s/it]
 99%|█████████▉| 14565/14725 [7:23:37<04:47,  1.80s/it]
 99%|█████████▉| 14566/14725 [7:23:39<04:45,  1.79s/it]
 99%|█████████▉| 14567/14725 [7:23:40<04:43,  1.79s/it]
 99%|█████████▉| 14568/14725 [7:23:42<04:42,  1.80s/it]
 99%|█████████▉| 14569/14725 [7:23:44<04:40,  1.80s/it]
 99%|█████████▉| 14570/14725 [7:23:46<04:39,  1.80s/it]
                                                       

 99%|█████████▉| 14570/14725 [7:23:46<04:39,  1.80s/it]
 99%|█████████▉| 14571/14725 [7:23:47<04:35,  1.79s/it]
 99%|█████████▉| 14572/14725 [7:23:49<04:40,  1.83s/it]
 99%|█████████▉| 14573/14725 [7:23
+0: {'loss': 0.7766, 'grad_norm': 0.6216361194949082, 'learning_rate': 1.0623119116437207e-05, 'epoch': 0.99}
+0: :51<04:33,  1.80s/it]
 99%|█████████▉| 14574/14725 [7:23:53<04:35,  1.82s/it]
 99%|█████████▉| 14575/14725 [7:23:55<04:30,  1.81s/it]
 99%|█████████▉| 14576/14725 [7:23:57<04:27,  1.79s/it]
 99%|█████████▉| 14577/14725 [7:23:58<04:22,  1.78s/it]
 99%|█████████▉| 14578/14725 [7:24:00<04:18,  1.76s/it]
 99%|█████████▉| 14579/14725 [7:24:02<04:19,  1.78s/it]
 99%|█████████▉| 14580/14725 [7:24:04<04:18,  1.79s/it]
                                                       

 99%|█████████▉| 14580/14725 [7:24:04<04:18,  1.79s/it]
 99%|█████████▉| 14581/14725 [7:24:05<04:16,  1.78s/it]
 99%|█████████▉| 14582/14725 [7:24:07<04:15,  1.78s/it]
 99%|█████████▉| 14583/14725 [7:24:09<04:16,  1.81s/it]
 99%|█████████▉| 14584/14725 [7:24:11<04:11,  1.78s/it]
 99%|█████████�
+0: {'loss': 0.7905, 'grad_norm': 0.623164579110252, 'learning_rate': 9.685252742938298e-06, 'epoch': 0.99}
+0: ��| 14585/14725 [7:24:12<04:06,  1.76s/it]
 99%|█████████▉| 14586/14725 [7:24:14<04:05,  1.77s/it]
 99%|█████████▉| 14587/14725 [7:24:16<04:02,  1.76s/it]
 99%|█████████▉| 14588/14725 [7:24:18<04:00,  1.75s/it]
 99%|█████████▉| 14589/14725 [7:24:20<04:03,  1.79s/it]
 99%|█████████▉| 14590/14725 [7:24:21<03:59,  1.77s/it]
                                                       

 99%|█████████▉| 14590/14725 [7:24:21<03:59,  1.77s/it]
 99%|█████████▉| 14591/14725 [7:24:23<03:59,  1.79s/it]
 99%|█████████▉| 14592/14725 [7:24:25<03:58,  1.79s/it]
 99%|█████████▉| 14593/14725 [7:24:27<03:57,  1.80s/it]
 99%|█████████▉| 14594/14725 [7:24:29<03:57,  1.82s/it]
 99%|█████████▉| 14595/14725 [7:24:30<03:52,  1.79s/it]
 99%|█████████▉| 14596/14725 [7:24:32<03:49,  1.78s/it]
 99%|██�
+0: {'loss': 0.7968, 'grad_norm': 0.6112496480919049, 'learning_rate': 8.76179101551631e-06, 'epoch': 0.99}
+0: ��██████▉| 14597/14725 [7:24:34<03:53,  1.82s/it]
 99%|█████████▉| 14598/14725 [7:24:36<03:51,  1.82s/it]
 99%|█████████▉| 14599/14725 [7:24:38<03:50,  1.83s/it]
 99%|█████████▉| 14600/14725 [7:24:40<03:47,  1.82s/it]
                                                       

 99%|█████████▉| 14600/14725 [7:24:40<03:47,  1.82s/it]
 99%|█████████▉| 14601/14725 [7:24:41<03:44,  1.81s/it]
 99%|█████████▉| 14602/14725 [7:24:43<03:40,  1.79s/it]
 99%|█████████▉| 14603/14725 [7:24:45<03:39,  1.80s/it]
 99%|█████████▉| 14604/14725 [7:24:47<03:34,  1.77s/it]
 99%|█████████▉| 14605/14725 [7:24:48<03:35,  1.80s/it]
 99%|█████████▉| 14606/14725 [7:24:50<03:34,  1.80s/it]
 99%|█████████▉| 14607/14725 [7:24:52<03:33,  1.81s/it]
 99%|█████████▉| 14608/14725 [7:24:54<03:29,  1
+0: {'loss': 0.7721, 'grad_norm': 0.6117598003780316, 'learning_rate': 7.862851574103665e-06, 'epoch': 0.99}
+0: .79s/it]
 99%|█████████▉| 14609/14725 [7:24:56<03:26,  1.78s/it]
 99%|█████████▉| 14610/14725 [7:24:57<03:23,  1.77s/it]
                                                       

 99%|█████████▉| 14610/14725 [7:24:57<03:23,  1.77s/it]
 99%|█████████▉| 14611/14725 [7:24:59<03:22,  1.77s/it]
 99%|█████████▉| 14612/14725 [7:25:01<03:18,  1.76s/it]
 99%|█████████▉| 14613/14725 [7:25:03<03:26,  1.84s/it]
 99%|█████████▉| 14614/14725 [7:25:05<03:22,  1.82s/it]
 99%|█████████▉| 14615/14725 [7:25:06<03:20,  1.83s/it]
 99%|█████████▉| 14616/14725 [7:25:08<03:21,  1.85s/it]
 99%|█████████▉| 14617/14725 [7:25:10<03:16,  1.82s/it]
 99%|█████████▉| 14618/14725 [7:25:12<03:13,  1.81s/it]
 99%|█████████▉| 14619/14725 [7:25:14<03:12,  1.82s/it]
 99%|█████████▉| 14620/147
+0: {'loss': 0.7896, 'grad_norm': 0.6265873046050205, 'learning_rate': 6.998283387335656e-06, 'epoch': 0.99}
+0: {'loss': 0.797, 'grad_norm': 0.5881472912117872, 'learning_rate': 6.177558845189029e-06, 'epoch': 0.99}
+0: 25 [7:25:16<03:10,  1.81s/it]
                                                       

 99%|█████████▉| 14620/14725 [7:25:16<03:10,  1.81s/it]
 99%|█████████▉| 14621/14725 [7:25:17<03:07,  1.81s/it]
 99%|█████████▉| 14622/14725 [7:25:19<03:04,  1.80s/it]
 99%|█████████▉| 14623/14725 [7:25:21<03:02,  1.79s/it]
 99%|█████████▉| 14624/14725 [7:25:23<02:58,  1.77s/it]
 99%|█████████▉| 14625/14725 [7:25:24<02:59,  1.80s/it]
 99%|█████████▉| 14626/14725 [7:25:26<02:56,  1.78s/it]
 99%|█████████▉| 14627/14725 [7:25:28<02:55,  1.79s/it]
 99%|█████████▉| 14628/14725 [7:25:30<02:53,  1.79s/it]
 99%|█████████▉| 14629/14725 [7:25:32<02:50,  1.78s/it]
 99%|█████████▉| 14630/14725 [7:25:33<02:47,  1.76s/it]
                                                       

 99%|█████████▉| 14630/1
+0: {'loss': 0.7861, 'grad_norm': 0.5891440928616382, 'learning_rate': 5.4096699774952075e-06, 'epoch': 0.99}
+0: 4725 [7:25:33<02:47,  1.76s/it]
 99%|█████████▉| 14631/14725 [7:25:35<02:47,  1.78s/it]
 99%|█████████▉| 14632/14725 [7:25:37<02:44,  1.77s/it]
 99%|█████████▉| 14633/14725 [7:25:39<02:41,  1.76s/it]
 99%|█████████▉| 14634/14725 [7:25:40<02:38,  1.74s/it]
 99%|█████████▉| 14635/14725 [7:25:42<02:36,  1.74s/it]
 99%|█████████▉| 14636/14725 [7:25:44<02:34,  1.74s/it]
 99%|█████████▉| 14637/14725 [7:25:46<02:33,  1.75s/it]
 99%|█████████▉| 14638/14725 [7:25:47<02:33,  1.77s/it]
 99%|█████████▉| 14639/14725 [7:25:49<02:32,  1.77s/it]
 99%|█████████▉| 14640/14725 [7:25:51<02:29,  1.76s/it]
                                                       

 99%|█████████▉| 14640/14725 [7:25:51<02:29,  1.76s/it]
 99%|█████████▉| 14641/14725 [7:25:53<02:27,  1.75s/it]
 99%|██████
+0: {'loss': 0.784, 'grad_norm': 0.6013821127183802, 'learning_rate': 4.703029935379711e-06, 'epoch': 0.99}
+0: ███▉| 14642/14725 [7:25:54<02:25,  1.75s/it]
 99%|█████████▉| 14643/14725 [7:25:56<02:23,  1.75s/it]
 99%|█████████▉| 14644/14725 [7:25:58<02:21,  1.74s/it]
 99%|█████████▉| 14645/14725 [7:26:00<02:19,  1.74s/it]
 99%|█████████▉| 14646/14725 [7:26:01<02:19,  1.77s/it]
 99%|█████████▉| 14647/14725 [7:26:03<02:16,  1.75s/it]
 99%|█████████▉| 14648/14725 [7:26:05<02:15,  1.76s/it]
 99%|█████████▉| 14649/14725 [7:26:07<02:15,  1.78s/it]
 99%|█████████▉| 14650/14725 [7:26:08<02:13,  1.78s/it]
                                                       

 99%|█████████▉| 14650/14725 [7:26:08<02:13,  1.78s/it]
 99%|█████████▉| 14651/14725 [7:26:10<02:12,  1.80s/it]
100%|█████████▉| 14652/14725 [7:26:12<02:10,  1.79s/it]
100%|█████████▉| 14653/14725 [7:26:14<02:08,  1.79s/it]
10
+0: {'loss': 0.7724, 'grad_norm': 0.5743299240666354, 'learning_rate': 4.065380815017898e-06, 'epoch': 1.0}
+0: 0%|█████████▉| 14654/14725 [7:26:16<02:07,  1.80s/it]
100%|█████████▉| 14655/14725 [7:26:18<02:06,  1.80s/it]
100%|█████████▉| 14656/14725 [7:26:19<02:04,  1.80s/it]
100%|█████████▉| 14657/14725 [7:26:21<02:03,  1.81s/it]
100%|█████████▉| 14658/14725 [7:26:23<01:59,  1.79s/it]
100%|█████████▉| 14659/14725 [7:26:25<01:57,  1.78s/it]
100%|█████████▉| 14660/14725 [7:26:26<01:55,  1.78s/it]
                                                       

100%|█████████▉| 14660/14725 [7:26:26<01:55,  1.78s/it]
100%|█████████▉| 14661/14725 [7:26:28<01:52,  1.76s/it]
100%|█████████▉| 14662/14725 [7:26:30<01:50,  1.76s/it]
100%|█████████▉| 14663/14725 [7:26:32<01:50,  1.78s/it]
100%|█████████▉| 14664/14725 [7:26:34<01:48,  1.78s/it]
100%|█████████▉| 14665/14725 [7:26:35
+0: {'loss': 0.781, 'grad_norm': 0.591710659881034, 'learning_rate': 3.5037088336091054e-06, 'epoch': 1.0}
+0: <01:46,  1.78s/it]
100%|█████████▉| 14666/14725 [7:26:37<01:44,  1.78s/it]
100%|█████████▉| 14667/14725 [7:26:39<01:42,  1.77s/it]
100%|█████████▉| 14668/14725 [7:26:41<01:40,  1.77s/it]
100%|█████████▉| 14669/14725 [7:26:42<01:39,  1.78s/it]
100%|█████████▉| 14670/14725 [7:26:44<01:37,  1.77s/it]
                                                       

100%|█████████▉| 14670/14725 [7:26:44<01:37,  1.77s/it]
100%|█████████▉| 14671/14725 [7:26:46<01:35,  1.77s/it]
100%|█████████▉| 14672/14725 [7:26:48<01:33,  1.76s/it]
100%|█████████▉| 14673/14725 [7:26:49<01:32,  1.77s/it]
100%|█████████▉| 14674/14725 [7:26:51<01:30,  1.77s/it]
100%|█████████▉| 14675/14725 [7:26:53<01:28,  1.78s/it]
100%|█████████▉| 14676/14725 [7:26:55<01:28,  1.81s/it]
100%|█████████▉|
+0: {'loss': 0.7817, 'grad_norm': 0.5755063372833804, 'learning_rate': 3.0241677869190676e-06, 'epoch': 1.0}
+0:  14677/14725 [7:26:57<01:25,  1.79s/it]
100%|█████████▉| 14678/14725 [7:26:58<01:23,  1.78s/it]
100%|█████████▉| 14679/14725 [7:27:00<01:22,  1.80s/it]
100%|█████████▉| 14680/14725 [7:27:02<01:21,  1.80s/it]
                                                       

100%|█████████▉| 14680/14725 [7:27:02<01:21,  1.80s/it]
100%|█████████▉| 14681/14725 [7:27:04<01:18,  1.79s/it]
100%|█████████▉| 14682/14725 [7:27:06<01:17,  1.79s/it]
100%|█████████▉| 14683/14725 [7:27:07<01:15,  1.81s/it]
100%|█████████▉| 14684/14725 [7:27:09<01:13,  1.79s/it]
100%|█████████▉| 14685/14725 [7:27:11<01:11,  1.78s/it]
100%|█████████▉| 14686/14725 [7:27:13<01:09,  1.79s/it]
100%|█████████▉| 14687/14725 [7:27:15<01:08,  1.80s/it]
100%|█████████▉| 14688/14725 [7:27:16<01:06,  1.79s/it]
100%|███�
+0: {'loss': 0.7845, 'grad_norm': 0.5710481162556568, 'learning_rate': 2.632011627005738e-06, 'epoch': 1.0}
+0: ��█████▉| 14689/14725 [7:27:18<01:04,  1.78s/it]
100%|█████████▉| 14690/14725 [7:27:20<01:02,  1.78s/it]
                                                       

100%|█████████▉| 14690/14725 [7:27:20<01:02,  1.78s/it]
100%|█████████▉| 14691/14725 [7:27:22<01:00,  1.77s/it]
100%|█████████▉| 14692/14725 [7:27:23<00:58,  1.77s/it]
100%|█████████▉| 14693/14725 [7:27:25<00:56,  1.78s/it]
100%|█████████▉| 14694/14725 [7:27:27<00:55,  1.78s/it]
100%|█████████▉| 14695/14725 [7:27:29<00:52,  1.76s/it]
100%|█████████▉| 14696/14725 [7:27:30<00:51,  1.76s/it]
100%|█████████▉| 14697/14725 [7:27:32<00:49,  1.75s/it]
100%|█████████▉| 14698/14725 [7:27:34<00:49,  1.83s/it]
100%|█████████▉| 14699/14725 [7:27:36<00:47,  1.84s/it]
100%|█████████▉| 14700/14725 [7:27:38<00:45,  1.83
+0: {'loss': 0.7802, 'grad_norm': 0.5828886354436302, 'learning_rate': 2.331536898821077e-06, 'epoch': 1.0}
+0: {'loss': 0.794, 'grad_norm': 0.5632691812705731, 'learning_rate': 2.1260356663654562e-06, 'epoch': 1.0}
+0: s/it]
                                                       

100%|█████████▉| 14700/14725 [7:27:38<00:45,  1.83s/it]
100%|█████████▉| 14701/14725 [7:27:40<00:44,  1.84s/it]
100%|█████████▉| 14702/14725 [7:27:42<00:42,  1.83s/it]
100%|█████████▉| 14703/14725 [7:27:43<00:39,  1.81s/it]
100%|█████████▉| 14704/14725 [7:27:45<00:37,  1.81s/it]
100%|█████████▉| 14705/14725 [7:27:47<00:35,  1.80s/it]
100%|█████████▉| 14706/14725 [7:27:49<00:34,  1.80s/it]
100%|█████████▉| 14707/14725 [7:27:50<00:32,  1.79s/it]
100%|█████████▉| 14708/14725 [7:27:52<00:30,  1.79s/it]
100%|█████████▉| 14709/14725 [7:27:54<00:28,  1.80s/it]
100%|█████████▉| 14710/14725 [7:27:56<00:27,  1.84s/it]
                                                       

100%|███████���█▉| 14710/14725 [7:27:56<00:27,  1.
+0: {'loss': 0.7621, 'grad_norm': 0.5513813094604884, 'learning_rate': 2.017759444145556e-06, 'epoch': 1.0}
+0: 84s/it]
100%|█████████▉| 14711/14725 [7:27:58<00:25,  1.81s/it]
100%|█████████▉| 14712/14725 [7:28:00<00:23,  1.83s/it]
100%|█████████▉| 14713/14725 [7:28:01<00:21,  1.80s/it]
100%|█████████▉| 14714/14725 [7:28:03<00:19,  1.79s/it]
100%|█████████▉| 14715/14725 [7:28:05<00:17,  1.77s/it]
100%|█████████▉| 14716/14725 [7:28:07<00:16,  1.78s/it]
100%|█████████▉| 14717/14725 [7:28:09<00:14,  1.81s/it]
100%|█████████▉| 14718/14725 [7:28:10<00:12,  1.79s/it]
100%|█████████▉| 14719/14725 [7:28:12<00:10,  1.77s/it]
100%|█████████▉| 14720/14725 [7:28:14<00:08,  1.77s/it]
                                                       

100%|█████████▉| 14720/14725 [7:28:14<00:08,  1.77s/it]
100%|█████████▉| 14721/14725 [7:28:16<00:07,  1.80s/it]
100%|█████████▉| 14722/1472
+0: {'train_runtime': 26906.8012, 'train_samples_per_second': 8.756, 'train_steps_per_second': 0.547, 'train_loss': 0.8366555213604394, 'epoch': 1.0}
+0: 5 [7:28:17<00:05,  1.78s/it]
100%|█████████▉| 14723/14725 [7:28:19<00:03,  1.78s/it]
100%|█████████▉| 14724/14725 [7:28:21<00:01,  1.76s/it]
100%|██████████| 14725/14725 [7:28:23<00:00,  1.78s/it]
                                                       

100%|██████████| 14725/14725 [7:28:26<00:00,  1.78s/it]
100%|██████████| 14725/14725 [7:28:26<00:00,  1.83s/it]
+0: [2025-08-07 07:16:46,736] [INFO] [axolotl.train.save_trained_model:246] [PID:411490] [RANK:0] Training completed! Saving trained model to /lustre/fswork/projects/rech/dgo/udv55np/ift/Qwen3-235B-A22B/Qwen2.5-0.5B/1.
+0: [2025-08-07 07:16:48,531] [INFO] [axolotl.train.save_trained_model:331] [PID:411490] [RANK:0] Model successfully saved to /lustre/fswork/projects/rech/dgo/udv55np/ift/Qwen3-235B-A22B/Qwen2.5-0.5B/1