# Load model directly
from transformers import AutoTokenizer, AutoModelForSequenceClassification
tokenizer = AutoTokenizer.from_pretrained("Dharil/toxic-initial-training")
model = AutoModelForSequenceClassification.from_pretrained("Dharil/toxic-initial-training")Quick Links
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: 0.3031
- Accuracy: 64.2857
- Hamming Loss: 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: 0.001
- train_batch_size: 16
- eval_batch_size: 64
- 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
- lr_scheduler_warmup_steps: 500
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Hamming Loss |
|---|---|---|---|---|---|
| 0.2107 | 1.0 | 11 | 0.3031 | 64.2857 | 0.1111 |
| 0.1632 | 2.0 | 22 | 0.3285 | 54.7619 | 0.1349 |
| 0.1236 | 3.0 | 33 | 0.3710 | 64.2857 | 0.0952 |
Framework versions
- Transformers 4.48.3
- Pytorch 2.5.1+cu124
- Datasets 3.3.2
- Tokenizers 0.21.0
- Downloads last month
- 1
Model tree for Dharil/toxic-initial-training
Base model
unitary/toxic-bert
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="Dharil/toxic-initial-training")