fresh-2-layer-qasc5000-distill-of-fresh-2-layer-mmlu_EVAL_mmlu

This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 207.7788
• Accuracy: 0.3764

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	202.2836	0.24
No log	1.27	200	206.3151	0.34
No log	1.91	300	195.0245	0.35
No log	2.55	400	201.5064	0.388
98.9054	3.18	500	200.6433	0.386
98.9054	3.82	600	193.7714	0.388
98.9054	4.46	700	189.6684	0.39
98.9054	5.1	800	192.2281	0.378
98.9054	5.73	900	194.5975	0.388
23.6632	6.37	1000	199.0726	0.42
23.6632	7.01	1100	192.3244	0.402
23.6632	7.64	1200	186.5772	0.402
23.6632	8.28	1300	190.7620	0.402
23.6632	8.92	1400	198.1279	0.394
14.2685	9.55	1500	194.6297	0.366

Framework versions

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