distilbert-base-uncased-finetuned-quora-insincere

This model is a fine-tuned version of distilbert-base-uncased on quora-insincere dataset. It achieves the following results on the evaluation set:

• Loss: 0.0946
• Accuracy: 0.9676
• F1 Score: 0.7309

Model description

LABEL_0 = Sincere question LABEL_1 = Insincere question

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: 20
• eval_batch_size: 20
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.093	1.0	62807	0.0859	0.9644
0.0695	2.0	125614	0.0946	0.9676

Evaluation results

'eval_loss': 0.09461139887571335, 'eval_accuracy': 0.9676, 'eval_f1': 0.7308970099667774,

Framework versions

• Transformers 4.34.1
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1