Output_llama3_80-20

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: 0.6274
• Balanced Accuracy: 0.6563
• Accuracy: 0.6615

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	0.6857	0.5462	0.5625
No log	2.0	192	0.6684	0.5758	0.5833
No log	3.0	288	0.7167	0.6384	0.6198
No log	4.0	384	0.6335	0.6179	0.625
No log	5.0	480	0.6574	0.6297	0.5990
0.6776	6.0	576	0.6322	0.6168	0.625
0.6776	7.0	672	0.6374	0.6114	0.6094
0.6776	8.0	768	0.6261	0.6278	0.6354
0.6776	9.0	864	0.6289	0.6651	0.6406
0.6776	10.0	960	0.6082	0.6368	0.6406
0.5732	11.0	1056	0.6036	0.6553	0.6615
0.5732	12.0	1152	0.6445	0.6870	0.6510
0.5732	13.0	1248	0.6094	0.6833	0.6875
0.5732	14.0	1344	0.6104	0.6607	0.6667
0.5732	15.0	1440	0.6553	0.6960	0.6927
0.5144	16.0	1536	0.6262	0.6603	0.6510
0.5144	17.0	1632	0.6154	0.6619	0.6667
0.5144	18.0	1728	0.6210	0.6619	0.6667
0.5144	19.0	1824	0.6293	0.6716	0.6771
0.5144	20.0	1920	0.6274	0.6563	0.6615

Framework versions

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