gpt2-case-25

This model is a fine-tuned version of gpt2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3446
• Accuracy: 0.8667

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: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 224
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	44	0.5467	0.8067
No log	2.0	88	0.5013	0.8267
No log	3.0	132	0.3836	0.82
No log	4.0	176	0.3446	0.8667
No log	5.0	220	0.4319	0.82
No log	6.0	264	0.4025	0.82
No log	7.0	308	0.4042	0.8333
No log	8.0	352	0.4388	0.84
No log	9.0	396	0.4096	0.8467
No log	10.0	440	0.4015	0.8733
No log	11.0	484	0.4120	0.8867
0.2552	12.0	528	0.4691	0.8667
0.2552	13.0	572	0.4504	0.86
0.2552	14.0	616	0.4902	0.88
0.2552	15.0	660	0.4885	0.8733
0.2552	16.0	704	0.5220	0.88
0.2552	17.0	748	0.5162	0.8733
0.2552	18.0	792	0.5554	0.8733
0.2552	19.0	836	0.5926	0.8733
0.2552	20.0	880	0.5597	0.86
0.2552	21.0	924	0.5755	0.8733
0.2552	22.0	968	0.5596	0.8867
0.0947	23.0	1012	0.5752	0.88
0.0947	24.0	1056	0.5745	0.88
0.0947	25.0	1100	0.5722	0.88

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0