fresh-2-layer-medmcqa2000-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: 10.1529
• Accuracy: 0.5909

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
• training_steps: 5000

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.59	100	13.5993	0.3737
No log	3.17	200	12.1119	0.4596
No log	4.76	300	11.4081	0.4646
No log	6.35	400	10.7470	0.5152
2.9844	7.94	500	10.2091	0.5152
2.9844	9.52	600	11.1542	0.5505
2.9844	11.11	700	10.3355	0.5404
2.9844	12.7	800	10.1297	0.5404
2.9844	14.29	900	10.4198	0.5303
0.4746	15.87	1000	10.0845	0.5556
0.4746	17.46	1100	10.2199	0.5404
0.4746	19.05	1200	10.1049	0.5404
0.4746	20.63	1300	10.1543	0.5404
0.4746	22.22	1400	10.3127	0.5606
0.2243	23.81	1500	10.1529	0.5909
0.2243	25.4	1600	10.0761	0.5707
0.2243	26.98	1700	10.3999	0.5859
0.2243	28.57	1800	10.1831	0.5859
0.2243	30.16	1900	10.2053	0.5909
0.1395	31.75	2000	10.4780	0.5455

Framework versions

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