Llama3_8B_final_Task2_2.0

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: 0.5334
• Accuracy: 0.9129
• Precision: 0.9050
• Recall: 0.9231
• F1 score: 0.9140

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: 0.0001
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1 score
0.6105	0.2725	200	0.3962	0.8486	0.8793	0.8091	0.8427
0.4553	0.5450	400	0.3388	0.8757	0.8729	0.8803	0.8766
0.4318	0.8174	600	0.3273	0.89	0.9006	0.8775	0.8889
0.3155	1.0899	800	0.3807	0.89	0.9281	0.8462	0.8852
0.3279	1.3624	1000	0.3757	0.8943	0.9601	0.8234	0.8865
0.2451	1.6349	1200	0.3784	0.89	0.8683	0.9202	0.8935
0.2956	1.9074	1400	0.3187	0.9143	0.9318	0.8946	0.9128
0.2107	2.1798	1600	0.3999	0.89	0.8513	0.9459	0.8961
0.1744	2.4523	1800	0.6330	0.8857	0.9788	0.7892	0.8738
0.191	2.7248	2000	0.4101	0.91	0.9444	0.8718	0.9067
0.1378	2.9973	2200	0.4604	0.8957	0.8582	0.9487	0.9012
0.0703	3.2698	2400	0.4276	0.9	0.8958	0.9060	0.9008
0.0582	3.5422	2600	0.5431	0.9086	0.9527	0.8604	0.9042
0.0887	3.8147	2800	0.4993	0.9157	0.9534	0.8746	0.9123
0.0976	4.0872	3000	0.4540	0.9157	0.9269	0.9031	0.9149
0.0179	4.3597	3200	0.5068	0.92	0.9086	0.9345	0.9213
0.0277	4.6322	3400	0.5119	0.9157	0.9269	0.9031	0.9149
0.0231	4.9046	3600	0.5334	0.9129	0.9050	0.9231	0.9140

Framework versions

• PEFT 0.12.0
• Transformers 4.44.2
• Pytorch 2.4.0+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1