Model save
Browse files- README.md +58 -0
- all_results.json +8 -0
- generation_config.json +9 -0
- train_results.json +8 -0
- trainer_state.json +497 -0
README.md
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: Qwen/Qwen2.5-Math-7B-Instruct
|
| 3 |
+
library_name: transformers
|
| 4 |
+
model_name: Qwen2.5-Math-7B-LIMO
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
- trl
|
| 8 |
+
- sft
|
| 9 |
+
licence: license
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# Model Card for Qwen2.5-Math-7B-LIMO
|
| 13 |
+
|
| 14 |
+
This model is a fine-tuned version of [Qwen/Qwen2.5-Math-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-Math-7B-Instruct).
|
| 15 |
+
It has been trained using [TRL](https://github.com/huggingface/trl).
|
| 16 |
+
|
| 17 |
+
## Quick start
|
| 18 |
+
|
| 19 |
+
```python
|
| 20 |
+
from transformers import pipeline
|
| 21 |
+
|
| 22 |
+
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
|
| 23 |
+
generator = pipeline("text-generation", model="flyingbugs/Qwen2.5-Math-7B-LIMO", device="cuda")
|
| 24 |
+
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
|
| 25 |
+
print(output["generated_text"])
|
| 26 |
+
```
|
| 27 |
+
|
| 28 |
+
## Training procedure
|
| 29 |
+
|
| 30 |
+
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/jjh233/huggingface/runs/rnhxytu3)
|
| 31 |
+
|
| 32 |
+
|
| 33 |
+
This model was trained with SFT.
|
| 34 |
+
|
| 35 |
+
### Framework versions
|
| 36 |
+
|
| 37 |
+
- TRL: 0.16.0.dev0
|
| 38 |
+
- Transformers: 4.49.0
|
| 39 |
+
- Pytorch: 2.5.1
|
| 40 |
+
- Datasets: 3.3.2
|
| 41 |
+
- Tokenizers: 0.21.0
|
| 42 |
+
|
| 43 |
+
## Citations
|
| 44 |
+
|
| 45 |
+
|
| 46 |
+
|
| 47 |
+
Cite TRL as:
|
| 48 |
+
|
| 49 |
+
```bibtex
|
| 50 |
+
@misc{vonwerra2022trl,
|
| 51 |
+
title = {{TRL: Transformer Reinforcement Learning}},
|
| 52 |
+
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
|
| 53 |
+
year = 2020,
|
| 54 |
+
journal = {GitHub repository},
|
| 55 |
+
publisher = {GitHub},
|
| 56 |
+
howpublished = {\url{https://github.com/huggingface/trl}}
|
| 57 |
+
}
|
| 58 |
+
```
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"total_flos": 6.364145551230894e+17,
|
| 3 |
+
"train_loss": 1.0789901279486143,
|
| 4 |
+
"train_runtime": 4151.3547,
|
| 5 |
+
"train_samples": 817,
|
| 6 |
+
"train_samples_per_second": 0.247,
|
| 7 |
+
"train_steps_per_second": 0.016
|
| 8 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"eos_token_id": [
|
| 4 |
+
151645,
|
| 5 |
+
151643
|
| 6 |
+
],
|
| 7 |
+
"pad_token_id": 151643,
|
| 8 |
+
"transformers_version": "4.49.0"
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"total_flos": 6.364145551230894e+17,
|
| 3 |
+
"train_loss": 1.0789901279486143,
|
| 4 |
+
"train_runtime": 4151.3547,
|
| 5 |
+
"train_samples": 817,
|
| 6 |
+
"train_samples_per_second": 0.247,
|
| 7 |
+
"train_steps_per_second": 0.016
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,497 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 65,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.07692307692307693,
|
| 13 |
+
"grad_norm": 56.69625199712202,
|
| 14 |
+
"learning_rate": 7.142857142857143e-06,
|
| 15 |
+
"loss": 10.6721,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.15384615384615385,
|
| 20 |
+
"grad_norm": 55.20314035746543,
|
| 21 |
+
"learning_rate": 1.4285714285714285e-05,
|
| 22 |
+
"loss": 10.847,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.23076923076923078,
|
| 27 |
+
"grad_norm": 62.29911555217105,
|
| 28 |
+
"learning_rate": 2.1428571428571428e-05,
|
| 29 |
+
"loss": 10.0456,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.3076923076923077,
|
| 34 |
+
"grad_norm": 60.40866589945923,
|
| 35 |
+
"learning_rate": 2.857142857142857e-05,
|
| 36 |
+
"loss": 3.7411,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.38461538461538464,
|
| 41 |
+
"grad_norm": 5.296024443977825,
|
| 42 |
+
"learning_rate": 3.571428571428572e-05,
|
| 43 |
+
"loss": 1.3917,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.46153846153846156,
|
| 48 |
+
"grad_norm": 11.63499005319304,
|
| 49 |
+
"learning_rate": 4.2857142857142856e-05,
|
| 50 |
+
"loss": 1.1571,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.5384615384615384,
|
| 55 |
+
"grad_norm": 2.310497639622062,
|
| 56 |
+
"learning_rate": 5e-05,
|
| 57 |
+
"loss": 1.0452,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.6153846153846154,
|
| 62 |
+
"grad_norm": 1.6470093668171653,
|
| 63 |
+
"learning_rate": 4.913793103448276e-05,
|
| 64 |
+
"loss": 1.0123,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.6923076923076923,
|
| 69 |
+
"grad_norm": 2.026871155602432,
|
| 70 |
+
"learning_rate": 4.827586206896552e-05,
|
| 71 |
+
"loss": 0.8787,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.7692307692307693,
|
| 76 |
+
"grad_norm": 1.36441223391375,
|
| 77 |
+
"learning_rate": 4.741379310344828e-05,
|
| 78 |
+
"loss": 0.8673,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.8461538461538461,
|
| 83 |
+
"grad_norm": 1.64152552079298,
|
| 84 |
+
"learning_rate": 4.655172413793104e-05,
|
| 85 |
+
"loss": 0.7986,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.9230769230769231,
|
| 90 |
+
"grad_norm": 1.084291869382038,
|
| 91 |
+
"learning_rate": 4.5689655172413794e-05,
|
| 92 |
+
"loss": 0.7852,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 1.0,
|
| 97 |
+
"grad_norm": 0.8244322351661075,
|
| 98 |
+
"learning_rate": 4.482758620689655e-05,
|
| 99 |
+
"loss": 0.7835,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 1.0769230769230769,
|
| 104 |
+
"grad_norm": 0.8467579577588499,
|
| 105 |
+
"learning_rate": 4.396551724137931e-05,
|
| 106 |
+
"loss": 0.7841,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 1.1538461538461537,
|
| 111 |
+
"grad_norm": 1.0394065154633032,
|
| 112 |
+
"learning_rate": 4.3103448275862066e-05,
|
| 113 |
+
"loss": 0.6965,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 1.2307692307692308,
|
| 118 |
+
"grad_norm": 0.6303665611617261,
|
| 119 |
+
"learning_rate": 4.224137931034483e-05,
|
| 120 |
+
"loss": 0.692,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 1.3076923076923077,
|
| 125 |
+
"grad_norm": 0.6650023019130861,
|
| 126 |
+
"learning_rate": 4.1379310344827587e-05,
|
| 127 |
+
"loss": 0.7196,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.3846153846153846,
|
| 132 |
+
"grad_norm": 0.5849872999722573,
|
| 133 |
+
"learning_rate": 4.0517241379310344e-05,
|
| 134 |
+
"loss": 0.6763,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 1.4615384615384617,
|
| 139 |
+
"grad_norm": 0.6568623192873284,
|
| 140 |
+
"learning_rate": 3.965517241379311e-05,
|
| 141 |
+
"loss": 0.6003,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 1.5384615384615383,
|
| 146 |
+
"grad_norm": 0.6188950192790953,
|
| 147 |
+
"learning_rate": 3.8793103448275865e-05,
|
| 148 |
+
"loss": 0.698,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.6153846153846154,
|
| 153 |
+
"grad_norm": 0.43937943678839847,
|
| 154 |
+
"learning_rate": 3.793103448275862e-05,
|
| 155 |
+
"loss": 0.587,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 1.6923076923076923,
|
| 160 |
+
"grad_norm": 0.46861722120782073,
|
| 161 |
+
"learning_rate": 3.7068965517241385e-05,
|
| 162 |
+
"loss": 0.6261,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 1.7692307692307692,
|
| 167 |
+
"grad_norm": 0.5938781657836611,
|
| 168 |
+
"learning_rate": 3.620689655172414e-05,
|
| 169 |
+
"loss": 0.7064,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 1.8461538461538463,
|
| 174 |
+
"grad_norm": 0.40181396579561623,
|
| 175 |
+
"learning_rate": 3.53448275862069e-05,
|
| 176 |
+
"loss": 0.5901,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.9230769230769231,
|
| 181 |
+
"grad_norm": 0.41241038369811467,
|
| 182 |
+
"learning_rate": 3.4482758620689657e-05,
|
| 183 |
+
"loss": 0.6272,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 2.0,
|
| 188 |
+
"grad_norm": 0.5098875911597754,
|
| 189 |
+
"learning_rate": 3.3620689655172414e-05,
|
| 190 |
+
"loss": 0.6515,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 2.076923076923077,
|
| 195 |
+
"grad_norm": 0.4339068791945304,
|
| 196 |
+
"learning_rate": 3.275862068965517e-05,
|
| 197 |
+
"loss": 0.5543,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 2.1538461538461537,
|
| 202 |
+
"grad_norm": 0.5054494096764832,
|
| 203 |
+
"learning_rate": 3.1896551724137935e-05,
|
| 204 |
+
"loss": 0.5018,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 2.230769230769231,
|
| 209 |
+
"grad_norm": 0.4238540025684798,
|
| 210 |
+
"learning_rate": 3.103448275862069e-05,
|
| 211 |
+
"loss": 0.6447,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 2.3076923076923075,
|
| 216 |
+
"grad_norm": 0.338738261804734,
|
| 217 |
+
"learning_rate": 3.017241379310345e-05,
|
| 218 |
+
"loss": 0.5072,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 2.3846153846153846,
|
| 223 |
+
"grad_norm": 0.38342740921195684,
|
| 224 |
+
"learning_rate": 2.9310344827586206e-05,
|
| 225 |
+
"loss": 0.5681,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 2.4615384615384617,
|
| 230 |
+
"grad_norm": 0.36692692741060295,
|
| 231 |
+
"learning_rate": 2.844827586206897e-05,
|
| 232 |
+
"loss": 0.5098,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 2.5384615384615383,
|
| 237 |
+
"grad_norm": 0.3649135699524806,
|
| 238 |
+
"learning_rate": 2.7586206896551727e-05,
|
| 239 |
+
"loss": 0.5079,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 2.6153846153846154,
|
| 244 |
+
"grad_norm": 0.3761181644174631,
|
| 245 |
+
"learning_rate": 2.672413793103448e-05,
|
| 246 |
+
"loss": 0.495,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 2.6923076923076925,
|
| 251 |
+
"grad_norm": 0.3417797037871458,
|
| 252 |
+
"learning_rate": 2.5862068965517244e-05,
|
| 253 |
+
"loss": 0.4813,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 2.769230769230769,
|
| 258 |
+
"grad_norm": 0.40945158138753057,
|
| 259 |
+
"learning_rate": 2.5e-05,
|
| 260 |
+
"loss": 0.562,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 2.8461538461538463,
|
| 265 |
+
"grad_norm": 0.3094047547298436,
|
| 266 |
+
"learning_rate": 2.413793103448276e-05,
|
| 267 |
+
"loss": 0.5303,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 2.9230769230769234,
|
| 272 |
+
"grad_norm": 0.4296909197321008,
|
| 273 |
+
"learning_rate": 2.327586206896552e-05,
|
| 274 |
+
"loss": 0.5714,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 3.0,
|
| 279 |
+
"grad_norm": 0.3729941423109703,
|
| 280 |
+
"learning_rate": 2.2413793103448276e-05,
|
| 281 |
+
"loss": 0.4989,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 3.076923076923077,
|
| 286 |
+
"grad_norm": 0.32599584121247344,
|
| 287 |
+
"learning_rate": 2.1551724137931033e-05,
|
| 288 |
+
"loss": 0.4522,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 3.1538461538461537,
|
| 293 |
+
"grad_norm": 0.36372367238403003,
|
| 294 |
+
"learning_rate": 2.0689655172413793e-05,
|
| 295 |
+
"loss": 0.4778,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 3.230769230769231,
|
| 300 |
+
"grad_norm": 0.3972570731654902,
|
| 301 |
+
"learning_rate": 1.9827586206896554e-05,
|
| 302 |
+
"loss": 0.4447,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 3.3076923076923075,
|
| 307 |
+
"grad_norm": 0.4012637330387787,
|
| 308 |
+
"learning_rate": 1.896551724137931e-05,
|
| 309 |
+
"loss": 0.4127,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 3.3846153846153846,
|
| 314 |
+
"grad_norm": 0.3026808512455307,
|
| 315 |
+
"learning_rate": 1.810344827586207e-05,
|
| 316 |
+
"loss": 0.4089,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 3.4615384615384617,
|
| 321 |
+
"grad_norm": 0.3511308932223884,
|
| 322 |
+
"learning_rate": 1.7241379310344828e-05,
|
| 323 |
+
"loss": 0.4866,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 3.5384615384615383,
|
| 328 |
+
"grad_norm": 0.44325984348261416,
|
| 329 |
+
"learning_rate": 1.6379310344827585e-05,
|
| 330 |
+
"loss": 0.4831,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 3.6153846153846154,
|
| 335 |
+
"grad_norm": 0.3193474058304498,
|
| 336 |
+
"learning_rate": 1.5517241379310346e-05,
|
| 337 |
+
"loss": 0.4396,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 3.6923076923076925,
|
| 342 |
+
"grad_norm": 0.3366172034999065,
|
| 343 |
+
"learning_rate": 1.4655172413793103e-05,
|
| 344 |
+
"loss": 0.451,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 3.769230769230769,
|
| 349 |
+
"grad_norm": 0.3192724937483484,
|
| 350 |
+
"learning_rate": 1.3793103448275863e-05,
|
| 351 |
+
"loss": 0.3647,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 3.8461538461538463,
|
| 356 |
+
"grad_norm": 0.41141681100708327,
|
| 357 |
+
"learning_rate": 1.2931034482758622e-05,
|
| 358 |
+
"loss": 0.4031,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 3.9230769230769234,
|
| 363 |
+
"grad_norm": 0.32878139231790354,
|
| 364 |
+
"learning_rate": 1.206896551724138e-05,
|
| 365 |
+
"loss": 0.4331,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 4.0,
|
| 370 |
+
"grad_norm": 0.2514182705601677,
|
| 371 |
+
"learning_rate": 1.1206896551724138e-05,
|
| 372 |
+
"loss": 0.4041,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 4.076923076923077,
|
| 377 |
+
"grad_norm": 0.30360346561484663,
|
| 378 |
+
"learning_rate": 1.0344827586206897e-05,
|
| 379 |
+
"loss": 0.4296,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 4.153846153846154,
|
| 384 |
+
"grad_norm": 0.37730490335834466,
|
| 385 |
+
"learning_rate": 9.482758620689655e-06,
|
| 386 |
+
"loss": 0.3607,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 4.230769230769231,
|
| 391 |
+
"grad_norm": 0.26735234806360975,
|
| 392 |
+
"learning_rate": 8.620689655172414e-06,
|
| 393 |
+
"loss": 0.3805,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 4.3076923076923075,
|
| 398 |
+
"grad_norm": 0.2666130294253567,
|
| 399 |
+
"learning_rate": 7.758620689655173e-06,
|
| 400 |
+
"loss": 0.3682,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 4.384615384615385,
|
| 405 |
+
"grad_norm": 0.30258601083800113,
|
| 406 |
+
"learning_rate": 6.896551724137932e-06,
|
| 407 |
+
"loss": 0.3824,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 4.461538461538462,
|
| 412 |
+
"grad_norm": 0.3051670708166745,
|
| 413 |
+
"learning_rate": 6.03448275862069e-06,
|
| 414 |
+
"loss": 0.3327,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 4.538461538461538,
|
| 419 |
+
"grad_norm": 0.37192436505550563,
|
| 420 |
+
"learning_rate": 5.172413793103448e-06,
|
| 421 |
+
"loss": 0.4074,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 4.615384615384615,
|
| 426 |
+
"grad_norm": 0.26152453728943026,
|
| 427 |
+
"learning_rate": 4.310344827586207e-06,
|
| 428 |
+
"loss": 0.3992,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 4.6923076923076925,
|
| 433 |
+
"grad_norm": 0.25734721330237487,
|
| 434 |
+
"learning_rate": 3.448275862068966e-06,
|
| 435 |
+
"loss": 0.4249,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 4.769230769230769,
|
| 440 |
+
"grad_norm": 0.23907794412322864,
|
| 441 |
+
"learning_rate": 2.586206896551724e-06,
|
| 442 |
+
"loss": 0.3945,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 4.846153846153846,
|
| 447 |
+
"grad_norm": 0.2511817058634514,
|
| 448 |
+
"learning_rate": 1.724137931034483e-06,
|
| 449 |
+
"loss": 0.3262,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 4.923076923076923,
|
| 454 |
+
"grad_norm": 0.25996184570432856,
|
| 455 |
+
"learning_rate": 8.620689655172415e-07,
|
| 456 |
+
"loss": 0.3407,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 5.0,
|
| 461 |
+
"grad_norm": 0.25930270839584474,
|
| 462 |
+
"learning_rate": 0.0,
|
| 463 |
+
"loss": 0.3129,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 5.0,
|
| 468 |
+
"step": 65,
|
| 469 |
+
"total_flos": 6.364145551230894e+17,
|
| 470 |
+
"train_loss": 1.0789901279486143,
|
| 471 |
+
"train_runtime": 4151.3547,
|
| 472 |
+
"train_samples_per_second": 0.247,
|
| 473 |
+
"train_steps_per_second": 0.016
|
| 474 |
+
}
|
| 475 |
+
],
|
| 476 |
+
"logging_steps": 1,
|
| 477 |
+
"max_steps": 65,
|
| 478 |
+
"num_input_tokens_seen": 0,
|
| 479 |
+
"num_train_epochs": 5,
|
| 480 |
+
"save_steps": 500,
|
| 481 |
+
"stateful_callbacks": {
|
| 482 |
+
"TrainerControl": {
|
| 483 |
+
"args": {
|
| 484 |
+
"should_epoch_stop": false,
|
| 485 |
+
"should_evaluate": false,
|
| 486 |
+
"should_log": false,
|
| 487 |
+
"should_save": true,
|
| 488 |
+
"should_training_stop": true
|
| 489 |
+
},
|
| 490 |
+
"attributes": {}
|
| 491 |
+
}
|
| 492 |
+
},
|
| 493 |
+
"total_flos": 6.364145551230894e+17,
|
| 494 |
+
"train_batch_size": 1,
|
| 495 |
+
"trial_name": null,
|
| 496 |
+
"trial_params": null
|
| 497 |
+
}
|