Dharil/Combined-labelled-psudeo-labelled-data-toxic-bert

results

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

• Loss: 0.1082
• Accuracy: 80.4114
• Hamming Loss: 0.0464

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: 0.001
• train_batch_size: 32
• eval_batch_size: 128
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Hamming Loss
0.1217	1.0	140	0.1082	80.4114	0.0464
0.2059	2.0	280	0.2015	75.9392	0.0625
0.2193	3.0	420	0.2308	76.1181	0.0732
0.204	4.0	560	0.2173	76.1181	0.0732

Framework versions

• Transformers 4.46.3
• Pytorch 2.4.1+cu121
• Tokenizers 0.20.3