gpt2-finetuned-v4-seinfeld

This model is a fine-tuned version of gpt2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.6941

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 1000
• num_epochs: 15
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
3.1734	0.79	8	2.9417
3.1576	1.59	16	2.9239
3.1234	2.4	24	2.8920
3.0712	3.2	32	2.8634
2.9738	3.99	40	2.8342
2.9761	4.79	48	2.8069
2.9294	5.59	56	2.7844
2.9026	6.4	64	2.7665
2.8501	7.2	72	2.7544
2.7805	7.99	80	2.7398
2.7905	8.79	88	2.7293
2.7661	9.59	96	2.7204
2.7272	10.4	104	2.7131
2.7092	11.2	112	2.7056
2.6392	11.99	120	2.7010
2.6468	12.79	128	2.6961
2.6269	13.59	136	2.6899
2.5952	14.4	144	2.6874

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.1+cu116
• Datasets 2.10.0
• Tokenizers 0.13.2