gpt2-text-classification-v2

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

• Loss: 0.2002
• Accuracy: 0.9342
• F1: 0.9340
• Recall: 0.9314
• Precision: 0.9367

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 3e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 3
• total_train_batch_size: 96
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Accuracy	F1	Validation Loss	Precision	Recall
0.327	0.9974	260	0.8973	0.8929	0.2559	0.9333	0.8558
0.241	1.9987	521	0.919	0.9180	0.2039	0.9296	0.9066
0.244	3.0	782	0.9154	0.9192	0.2156	0.8799	0.9621
0.1843	3.9974	1042	0.9299	0.9288	0.1888	0.9427	0.9154
0.1608	4.9987	1303	0.9301	0.9291	0.1855	0.9428	0.9158
0.124	6.0	1564	0.9322	0.9319	0.1826	0.9357	0.9282
0.112	6.9974	1820	0.2099	0.9315	0.9303	0.9138	0.9473
0.0903	7.9987	2081	0.2002	0.9342	0.9340	0.9314	0.9367

Framework versions

• Transformers 4.41.0
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1