prompt-tackler
This model is a fine-tuned version of protectai/deberta-v3-small-prompt-injection-v2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0186
- Accuracy: 0.9959
- Precision: 0.9959
- Recall: 0.9959
- F1: 0.9959
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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- 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: 6
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.0177 | 1.0 | 20686 | 0.0222 | 0.9943 | 0.9943 | 0.9943 | 0.9943 |
| 0.012 | 2.0 | 41372 | 0.0186 | 0.9959 | 0.9959 | 0.9959 | 0.9959 |
| 0.0084 | 3.0 | 62058 | 0.0278 | 0.9955 | 0.9955 | 0.9955 | 0.9955 |
| 0.0216 | 4.0 | 82744 | 0.0256 | 0.9959 | 0.9959 | 0.9959 | 0.9959 |
| 0.0038 | 5.0 | 103430 | 0.0327 | 0.9963 | 0.9963 | 0.9963 | 0.9963 |
| 0.0 | 6.0 | 124116 | 0.0383 | 0.9963 | 0.9963 | 0.9963 | 0.9963 |
Framework versions
- Transformers 4.53.3
- Pytorch 2.9.1+cu128
- Datasets 2.21.0
- Tokenizers 0.21.4