train_wsc_42_1760620823

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

• Loss: 0.3719
• Num Input Tokens Seen: 1308280

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: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Input Tokens Seen
0.5173	3.0	333	0.4170	130808
0.3604	6.0	666	0.3643	260688
0.3618	9.0	999	0.3509	390936
0.3679	12.0	1332	0.3535	521840
0.3368	15.0	1665	0.3609	653216
0.3391	18.0	1998	0.3556	784328
0.3735	21.0	2331	0.3569	916192
0.3254	24.0	2664	0.3660	1046952
0.3343	27.0	2997	0.3723	1177504
0.3324	30.0	3330	0.3719	1308280

Framework versions

• PEFT 0.15.2
• Transformers 4.51.3
• Pytorch 2.8.0+cu128
• Datasets 3.6.0
• Tokenizers 0.21.1