train_sst2_42_1760637618

This model is a fine-tuned version of meta-llama/Meta-Llama-3-8B-Instruct on the sst2 dataset. It achieves the following results on the evaluation set:

• Loss: 0.2804
• Num Input Tokens Seen: 60199256

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: 4
• eval_batch_size: 4
• 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: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Input Tokens Seen
0.0238	2.0	26940	0.0787	6016976
0.0747	4.0	53880	0.0791	12034136
0.055	6.0	80820	0.0902	18055800
0.0008	8.0	107760	0.1106	24070232
0.122	10.0	134700	0.1308	30091496
0.0	12.0	161640	0.1617	36113136
0.0	14.0	188580	0.1783	42129840
0.0	16.0	215520	0.2557	48154480
0.0	18.0	242460	0.2720	54176896
0.0	20.0	269400	0.2804	60199256

Framework versions

• PEFT 0.17.1
• Transformers 4.51.3
• Pytorch 2.9.0+cu128
• Datasets 4.0.0
• Tokenizers 0.21.4