fresh-2-layer-piqa5000-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: 13.3644
• Accuracy: 0.4444

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	0.64	100	16.9077	0.2677
No log	1.27	200	16.7142	0.3333
No log	1.91	300	13.9095	0.3131
No log	2.55	400	16.4004	0.3838
2.6921	3.18	500	13.3775	0.3939
2.6921	3.82	600	15.2723	0.4091
2.6921	4.46	700	13.8835	0.3939
2.6921	5.1	800	13.7036	0.4242
2.6921	5.73	900	14.1660	0.4293
0.7686	6.37	1000	15.2431	0.4394
0.7686	7.01	1100	14.4889	0.3838
0.7686	7.64	1200	14.0171	0.3889
0.7686	8.28	1300	13.3644	0.4444
0.7686	8.92	1400	13.8387	0.4192
0.3654	9.55	1500	14.4017	0.3838
0.3654	10.19	1600	13.9023	0.4293
0.3654	10.83	1700	15.1041	0.4091
0.3654	11.46	1800	14.6756	0.4141

Framework versions

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