Output_llama3_80-20_New_balanced

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: 0.7262
• Balanced Accuracy: 0.6223
• Accuracy: 0.6196

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	68	0.7155	0.4641	0.4641
No log	2.0	136	0.6814	0.5713	0.5694
No log	3.0	204	0.6994	0.6056	0.5742
No log	4.0	272	0.6697	0.5995	0.5957
No log	5.0	340	0.6533	0.5965	0.5957
No log	6.0	408	0.7671	0.6170	0.5622
No log	7.0	476	0.6653	0.6022	0.6005
0.7416	8.0	544	0.6865	0.6115	0.6029
0.7416	9.0	612	0.6997	0.6185	0.6029
0.7416	10.0	680	0.7029	0.6115	0.6029
0.7416	11.0	748	0.6979	0.5995	0.5957
0.7416	12.0	816	0.7114	0.6008	0.5957
0.7416	13.0	884	0.6943	0.6142	0.6124
0.7416	14.0	952	0.6944	0.6214	0.6196
0.5388	15.0	1020	0.6719	0.6293	0.6292
0.5388	16.0	1088	0.6958	0.6061	0.6053
0.5388	17.0	1156	0.6904	0.6175	0.6172
0.5388	18.0	1224	0.7013	0.6254	0.6244
0.5388	19.0	1292	0.7266	0.6223	0.6196
0.5388	20.0	1360	0.7262	0.6223	0.6196

Framework versions

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