fresh-2-layer-qasc5000-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: 14.9541
• Accuracy: 0.4949

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	14.0782	0.3131
No log	1.27	200	15.6534	0.4596
No log	1.91	300	16.2410	0.4293
No log	2.55	400	14.4394	0.4495
1.861	3.18	500	15.1088	0.4646
1.861	3.82	600	14.4621	0.4848
1.861	4.46	700	15.0037	0.4596
1.861	5.1	800	15.6656	0.4293
1.861	5.73	900	15.1955	0.4343
0.2711	6.37	1000	14.9541	0.4949
0.2711	7.01	1100	14.5891	0.4646
0.2711	7.64	1200	15.4136	0.4747
0.2711	8.28	1300	15.4486	0.4697
0.2711	8.92	1400	14.5015	0.4848
0.1568	9.55	1500	14.2260	0.4545

Framework versions

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