train_sst2_456_1760637847

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: 0.2714
• Num Input Tokens Seen: 60239960

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: 456
• 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
0.0265	2.0	26940	0.1004	6026120
0.1113	4.0	53880	0.0876	12045224
0.0359	6.0	80820	0.0993	18069072
0.0003	8.0	107760	0.0990	24094184
0.0006	10.0	134700	0.1489	30117800
0.1879	12.0	161640	0.1370	36146432
0.0008	14.0	188580	0.1625	42170168
0.0	16.0	215520	0.2505	48194664
0.0	18.0	242460	0.2595	54218504
0.0	20.0	269400	0.2714	60239960

Framework versions

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