fresh-2-layer-swag10000-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: 14.9335
• Accuracy: 0.4646

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	0.32	100	15.8202	0.2778
No log	0.64	200	14.7041	0.3384
No log	0.96	300	16.9031	0.3737
No log	1.28	400	18.0978	0.4141
2.1655	1.6	500	16.5271	0.4040
2.1655	1.92	600	14.4014	0.3990
2.1655	2.24	700	19.0358	0.4242
2.1655	2.56	800	14.9314	0.4192
2.1655	2.88	900	14.9335	0.4646
0.5334	3.19	1000	15.1769	0.4596
0.5334	3.51	1100	15.4032	0.4343
0.5334	3.83	1200	13.1365	0.4646
0.5334	4.15	1300	12.7464	0.4394
0.5334	4.47	1400	13.5877	0.4545

Framework versions

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