sentimental-gpt

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

• Loss: 0.5305
• Accuracy: 0.7985

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.7515	0.2910	500	0.6684	0.7187
0.6015	0.5821	1000	0.5922	0.7609
0.5758	0.8731	1500	0.5355	0.7804
0.5301	1.1641	2000	0.5441	0.7801
0.4545	1.4552	2500	0.5318	0.7917
0.5103	1.7462	3000	0.5479	0.7796
0.4495	2.0373	3500	0.5212	0.7906
0.4506	2.3283	4000	0.5282	0.7963
0.4271	2.6193	4500	0.5305	0.7985
0.44	2.9104	5000	0.5335	0.7971

Framework versions

• Transformers 4.49.0
• Pytorch 2.0.0+cu118
• Datasets 3.3.1
• Tokenizers 0.21.0