fresh-2-layer-medmcqa20000-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: 5.6554
• Accuracy: 0.7323

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	0.16	100	15.6094	0.3384
No log	0.32	200	11.2486	0.4747
No log	0.48	300	9.7003	0.5556
No log	0.64	400	9.0128	0.6212
3.508	0.8	500	9.0146	0.6111
3.508	0.96	600	7.8672	0.6465
3.508	1.12	700	7.4123	0.6768
3.508	1.28	800	7.4161	0.6515
3.508	1.44	900	7.3679	0.6667
0.9934	1.6	1000	6.8781	0.6768
0.9934	1.76	1100	6.7447	0.6717
0.9934	1.92	1200	7.6242	0.6818
0.9934	2.08	1300	6.6278	0.6869
0.9934	2.24	1400	6.4569	0.6970
0.6102	2.4	1500	6.7556	0.7121
0.6102	2.56	1600	6.3065	0.7020
0.6102	2.72	1700	6.2906	0.7071
0.6102	2.88	1800	6.2230	0.7121
0.6102	3.04	1900	6.9069	0.7172
0.4278	3.2	2000	6.0706	0.7172
0.4278	3.36	2100	6.0340	0.7172
0.4278	3.52	2200	5.6554	0.7323
0.4278	3.68	2300	5.8829	0.7020
0.4278	3.84	2400	5.6822	0.7222
0.3174	4.0	2500	5.7620	0.7222
0.3174	4.16	2600	5.7952	0.7172
0.3174	4.32	2700	5.6193	0.7222

Framework versions

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