train_openbookqa_456_1760637796

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

• Loss: 1.5929
• Num Input Tokens Seen: 7547264

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
0.6714	2.0	1984	0.7040	755080
0.6782	4.0	3968	0.6916	1511048
0.6286	6.0	5952	0.5775	2266184
0.313	8.0	7936	0.5477	3019344
0.45	10.0	9920	0.5980	3774832
0.6855	12.0	11904	0.6951	4529256
0.167	14.0	13888	0.8250	5283808
0.703	16.0	15872	1.2808	6038696
0.1082	18.0	17856	1.5583	6792504
0.1054	20.0	19840	1.5929	7547264

Framework versions

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