llama-3.1-phishing-adapter-a100

This model is a fine-tuned version of meta-llama/Llama-3.1-8B-Instruct on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0060
• Accuracy: 1.0

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0002
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.PAGED_ADAMW with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.03
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.2262	0.0870	25	0.0661	1.0
0.0715	0.1739	50	0.0525	0.9444
0.0725	0.2609	75	0.0136	0.5500
0.0252	0.3478	100	0.0190	0.8662
0.0109	0.4348	125	0.0392	0.9887
0.0315	0.5217	150	0.0529	1.0
0.0123	0.6087	175	0.0195	1.0
0.0124	0.6957	200	0.0102	1.0
0.0139	0.7826	225	0.0056	1.0
0.0124	0.8696	250	0.0065	1.0
0.0111	0.9565	275	0.0060	1.0

Framework versions

• PEFT 0.18.1
• Transformers 4.57.6
• Pytorch 2.9.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.2