fresh-2-layer-swag50000-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: 11.8130
• Accuracy: 0.5404

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	0.06	100	14.8090	0.2525
No log	0.13	200	21.7092	0.3636
No log	0.19	300	18.5664	0.3586
No log	0.26	400	17.2545	0.4091
2.298	0.32	500	14.6083	0.4646
2.298	0.38	600	16.9225	0.4697
2.298	0.45	700	14.4167	0.4747
2.298	0.51	800	14.2497	0.4949
2.298	0.58	900	12.1858	0.5202
0.6557	0.64	1000	12.1346	0.5101
0.6557	0.7	1100	12.0078	0.5101
0.6557	0.77	1200	11.9163	0.5354
0.6557	0.83	1300	13.2215	0.4899
0.6557	0.9	1400	11.8032	0.5051
0.4142	0.96	1500	11.8130	0.5404
0.4142	1.02	1600	11.6673	0.5253
0.4142	1.09	1700	11.6830	0.4798
0.4142	1.15	1800	11.5380	0.4949
0.4142	1.22	1900	12.0938	0.5101
0.2783	1.28	2000	11.6623	0.5202

Framework versions

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