facebook-hate-speech-fine-tuned

This model is a fine-tuned version of facebook/roberta-hate-speech-dynabench-r4-target on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1943
• Accuracy: 0.9636
• Precision Macro: 0.8675
• Recall Macro: 0.8731
• F1 Macro: 0.8703
• Precision Micro: 0.9636
• Recall Micro: 0.9636
• F1 Micro: 0.9636

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision Macro	Recall Macro	F1 Macro	Precision Micro	Recall Micro	F1 Micro
0.2628	1.0	199	0.1203	0.9585	0.8840	0.7939	0.8318	0.9585	0.9585	0.9585
0.071	2.0	398	0.1640	0.9673	0.9144	0.8369	0.8709	0.9673	0.9673	0.9673
0.1483	3.0	597	0.1943	0.9636	0.8675	0.8731	0.8703	0.9636	0.9636	0.9636

Framework versions

• Transformers 4.42.4
• Pytorch 2.4.0+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1