train_sst2_101112_1760638074

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: 0.2894
• Num Input Tokens Seen: 60228184

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: 101112
• 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
0.1226	2.0	26940	0.0985	6023008
0.2208	4.0	53880	0.0855	12044416
0.0037	6.0	80820	0.0970	18070520
0.0003	8.0	107760	0.1072	24093464
0.0003	10.0	134700	0.1501	30113600
0.0	12.0	161640	0.1856	36140272
0.0	14.0	188580	0.1940	42163056
0.0	16.0	215520	0.2507	48188240
0.0	18.0	242460	0.2721	54207032
0.0	20.0	269400	0.2894	60228184

Framework versions

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