fresh-2-layer-arc2000-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: 13.5990
• Accuracy: 0.4596

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	1.59	100	14.0069	0.2828
No log	3.17	200	15.3225	0.4040
No log	4.76	300	13.6218	0.4040
No log	6.35	400	15.3356	0.4242
2.6428	7.94	500	13.0031	0.4293
2.6428	9.52	600	13.0853	0.4343
2.6428	11.11	700	13.6346	0.4545
2.6428	12.7	800	13.1762	0.4394
2.6428	14.29	900	12.9734	0.4545
0.4544	15.87	1000	13.5990	0.4596
0.4544	17.46	1100	13.0461	0.4596
0.4544	19.05	1200	12.0434	0.4495
0.4544	20.63	1300	12.9092	0.4343
0.4544	22.22	1400	12.1925	0.4596
0.2055	23.81	1500	13.0411	0.4242

Framework versions

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