gpt2-finetuned-mcqa-sciq

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

• Loss: 2.3533
• Bertscore Precision: 0.1082
• Bertscore Recall: 0.1141
• Bertscore F1: 0.1111

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: 3e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• 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	Validation Loss	Bertscore Precision	Bertscore Recall	Bertscore F1
4.4695	0.9999	5839	2.3612	0.1082	0.1140	0.1110
4.0507	2.0	11679	2.3533	0.1082	0.1141	0.1111
3.8779	2.9999	17518	2.3820	0.1080	0.1140	0.1110
3.2852	4.0	23358	2.4208	0.1080	0.1140	0.1109
3.6416	4.9999	29197	2.4768	0.1079	0.1139	0.1108
2.9843	6.0	35037	2.5445	0.1079	0.1139	0.1108
2.8509	6.9999	40876	2.6094	0.1079	0.1139	0.1108
2.6932	8.0	46716	2.6658	0.1078	0.1138	0.1107
2.5309	8.9999	52555	2.7283	0.1078	0.1138	0.1107
2.5619	9.9991	58390	2.7585	0.1078	0.1138	0.1107

Framework versions

• Transformers 4.40.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1