gte-large-en-v1.5-based-ft-prompt-injection-detection-241205Weighted-44
This model is a fine-tuned version of Alibaba-NLP/gte-large-en-v1.5 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2193
- F1: 0.9463
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: 0.0001
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 256
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1000
- num_epochs: 50
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 |
|---|---|---|---|---|
| 0.4603 | 0.2527 | 100 | 0.2243 | 0.9083 |
| 0.2014 | 0.5054 | 200 | 0.1719 | 0.9377 |
| 0.1679 | 0.7581 | 300 | 0.1354 | 0.9486 |
| 0.1497 | 1.0107 | 400 | 0.1174 | 0.9543 |
| 0.1062 | 1.2634 | 500 | 0.1451 | 0.9490 |
| 0.1117 | 1.5161 | 600 | 0.1379 | 0.9504 |
| 0.1156 | 1.7688 | 700 | 0.1235 | 0.9533 |
| 0.1155 | 2.0215 | 800 | 0.1305 | 0.9562 |
| 0.0784 | 2.2742 | 900 | 0.1372 | 0.9530 |
| 0.0898 | 2.5268 | 1000 | 0.1204 | 0.9585 |
| 0.0904 | 2.7795 | 1100 | 0.1864 | 0.9294 |
| 0.0936 | 3.0322 | 1200 | 0.1731 | 0.9543 |
| 0.0644 | 3.2849 | 1300 | 0.1482 | 0.95 |
| 0.0668 | 3.5376 | 1400 | 0.1713 | 0.9455 |
| 0.0648 | 3.7903 | 1500 | 0.1499 | 0.9551 |
| 0.0605 | 4.0430 | 1600 | 0.2150 | 0.9506 |
| 0.0468 | 4.2956 | 1700 | 0.1913 | 0.9453 |
| 0.0531 | 4.5483 | 1800 | 0.2022 | 0.9421 |
| 0.0564 | 4.8010 | 1900 | 0.1694 | 0.9544 |
| 0.0568 | 5.0537 | 2000 | 0.2193 | 0.9463 |
Framework versions
- Transformers 4.45.2
- Pytorch 2.5.1+cu124
- Datasets 3.1.0
- Tokenizers 0.20.3
- Downloads last month
- 10
Inference API (serverless) does not yet support model repos that contain custom code.
Model tree for J1N2/gte-large-en-v1.5-based-ft-prompt-injection-detection-241205Weighted-44
Base model
Alibaba-NLP/gte-large-en-v1.5