fresh-2-layer-swag20000-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: 15.3688
• 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: 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	14.6793	0.2677
No log	0.32	200	21.2273	0.3687
No log	0.48	300	19.4173	0.3485
No log	0.64	400	23.6541	0.4343
2.2803	0.8	500	18.0346	0.4091
2.2803	0.96	600	20.1434	0.4293
2.2803	1.12	700	16.6838	0.4192
2.2803	1.28	800	15.3688	0.4747
2.2803	1.44	900	16.6134	0.4444
0.5776	1.6	1000	15.2527	0.4192
0.5776	1.76	1100	13.8964	0.4444
0.5776	1.92	1200	15.0290	0.4444
0.5776	2.08	1300	14.4659	0.4495

Framework versions

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