fresh-2-layer-medmcqa-distill-of-fresh-2-layer-gpqa-loop-5

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

• Loss: 0.4393
• Accuracy: 0.5202

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	2.3938	0.2475
No log	2.0	126	1.5164	0.3636
No log	3.0	189	1.1653	0.4646
No log	4.0	252	0.7958	0.4394
No log	5.0	315	0.5525	0.4596
No log	6.0	378	1.1572	0.4747
No log	7.0	441	0.3450	0.4798
1.7802	8.0	504	0.4393	0.5202
1.7802	9.0	567	0.5459	0.4343
1.7802	10.0	630	0.4935	0.5101
1.7802	11.0	693	0.3405	0.4697
1.7802	12.0	756	0.3275	0.4697
1.7802	13.0	819	0.2442	0.4646
1.7802	14.0	882	0.2561	0.4495
1.7802	15.0	945	0.2196	0.4495
0.215	16.0	1008	0.1943	0.4495
0.215	17.0	1071	0.1845	0.4545
0.215	18.0	1134	0.1702	0.4444
0.215	19.0	1197	0.1788	0.4545
0.215	20.0	1260	0.1747	0.4545

Framework versions

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