multilingual-toxic-comment-classifier

This model is a fine-tuned version of bert-base-multilingual-cased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0809
• Subset Accuracy: 0.91
• F1 Micro: 0.6923
• Precision Micro: 0.8182
• Recall Micro: 0.6
• Optimal Threshold: 0.3700

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: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Subset Accuracy	F1 Micro	Precision Micro	Recall Micro	Optimal Threshold
0.1653	1.0	100	0.1619	0.895	0.0	0.0	0.0	0.5
0.15	2.0	200	0.1527	0.895	0.1633	1.0	0.0889	0.3
0.0722	3.0	300	0.0809	0.91	0.6923	0.8182	0.6	0.3700

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1