Output_llama3_80-20_New

This model is a fine-tuned version of meta-llama/Meta-Llama-3-8B on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.6519
• Balanced Accuracy: 0.6770
• Accuracy: 0.7188

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: 1e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Balanced Accuracy	Accuracy
No log	1.0	96	0.6701	0.5509	0.5
No log	2.0	192	0.5817	0.7730	0.7865
No log	3.0	288	0.5816	0.6611	0.6719
No log	4.0	384	0.5791	0.6917	0.7344
No log	5.0	480	0.6088	0.7279	0.7604
0.6212	6.0	576	0.5876	0.6647	0.6771
0.6212	7.0	672	0.6213	0.6351	0.5938
0.6212	8.0	768	0.5972	0.6976	0.7396
0.6212	9.0	864	0.6078	0.6858	0.7292
0.6212	10.0	960	0.6185	0.6552	0.6771
0.5248	11.0	1056	0.6358	0.6526	0.6875
0.5248	12.0	1152	0.6519	0.6770	0.7188

Framework versions

• PEFT 0.10.0
• Transformers 4.46.3
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.20.3