fresh-2-layer-medmcqa-distill-of-fresh-2-layer-gpqa-loop-6

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

• Loss: 0.4081
• Accuracy: 0.4747

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: 16
• eval_batch_size: 16
• 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	63	2.4015	0.2879
No log	2.0	126	1.5345	0.3535
No log	3.0	189	0.8667	0.4091
No log	4.0	252	1.4331	0.4293
No log	5.0	315	0.5196	0.4293
No log	6.0	378	1.4378	0.4293
No log	7.0	441	0.5950	0.4343
1.7735	8.0	504	1.0409	0.4192
1.7735	9.0	567	0.7609	0.4394
1.7735	10.0	630	0.4812	0.4596
1.7735	11.0	693	0.5484	0.4394
1.7735	12.0	756	0.7745	0.4495
1.7735	13.0	819	0.3796	0.4545
1.7735	14.0	882	0.3806	0.4545
1.7735	15.0	945	0.4081	0.4747
0.2002	16.0	1008	0.3168	0.4596
0.2002	17.0	1071	0.3836	0.4545
0.2002	18.0	1134	0.3982	0.4646
0.2002	19.0	1197	0.2785	0.4697
0.2002	20.0	1260	0.3001	0.4697

Framework versions

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