fresh-2-layer-medmcqa-distill-of-fresh-2-layer-gpqa_EVAL_gpqa

This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 8.0989
• Accuracy: 0.6566

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: 0.0005
• train_batch_size: 32
• eval_batch_size: 32
• seed: 321
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	125	13.0027	0.4747
No log	2.0	250	10.8183	0.5404
No log	3.0	375	11.0325	0.5909
3.3093	4.0	500	11.0605	0.5808
3.3093	5.0	625	9.5436	0.5758
3.3093	6.0	750	9.1106	0.6515
3.3093	7.0	875	8.4697	0.6212
0.675	8.0	1000	9.1724	0.6212
0.675	9.0	1125	8.4508	0.6515
0.675	10.0	1250	8.5147	0.6111
0.675	11.0	1375	8.4648	0.6414
0.2645	12.0	1500	8.2626	0.6515
0.2645	13.0	1625	8.2865	0.6515
0.2645	14.0	1750	8.1180	0.6465
0.2645	15.0	1875	8.5052	0.6414
0.1402	16.0	2000	7.9762	0.6515
0.1402	17.0	2125	8.1063	0.6515
0.1402	18.0	2250	8.0695	0.6515
0.1402	19.0	2375	8.0989	0.6566
0.07	20.0	2500	8.0972	0.6566

Framework versions

• Transformers 4.34.0.dev0
• Pytorch 2.0.1+cu117
• Datasets 2.14.5
• Tokenizers 0.14.0