train_sst2_123_1760637733

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: 0.2979
• Num Input Tokens Seen: 60239952

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: 123
• 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
0.1249	2.0	26940	0.0897	6019008
0.076	4.0	53880	0.0957	12043464
0.0563	6.0	80820	0.0787	18067712
0.0003	8.0	107760	0.1176	24092912
0.0002	10.0	134700	0.1384	30119504
0.0	12.0	161640	0.1610	36141312
0.0	14.0	188580	0.2169	42169824
0.0001	16.0	215520	0.2601	48195672
0.0	18.0	242460	0.2935	54219304
0.0	20.0	269400	0.2979	60239952

Framework versions

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