fresh-2-layer-swag2000-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: 13.0130
• Accuracy: 0.4293

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	14.3215	0.2778
No log	3.17	200	17.1021	0.3687
No log	4.76	300	15.4345	0.3939
No log	6.35	400	14.2605	0.3939
1.9591	7.94	500	17.3736	0.3889
1.9591	9.52	600	14.1689	0.3939
1.9591	11.11	700	14.8678	0.3889
1.9591	12.7	800	13.5466	0.4040
1.9591	14.29	900	14.6713	0.4040
0.3021	15.87	1000	14.0434	0.4091
0.3021	17.46	1100	14.3483	0.4091
0.3021	19.05	1200	13.0130	0.4293
0.3021	20.63	1300	13.0252	0.3939
0.3021	22.22	1400	13.8034	0.3838
0.1527	23.81	1500	13.7069	0.4141
0.1527	25.4	1600	13.5824	0.3939
0.1527	26.98	1700	14.0297	0.3889

Framework versions

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