metadata
license: mit
base_model: microsoft/deberta-v3-base
tags:
- generated_from_trainer
- prompt injection
- security
- jailbreak
- prompt security
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: prompt-tackler
results: []
datasets:
- reshabhs/SPML_Chatbot_Prompt_Injection
- VMware/open-instruct
- jackhhao/jailbreak-classification
- cgoosen/prompt_injection_combined
language:
- en
- afr
- fr
library_name: transformers
prompt-tackler
This model is a fine-tuned version of microsoft/deberta-v3-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0101
- Accuracy: 0.9984
- Precision: 0.9984
- Recall: 0.9984
- F1: 0.9984
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
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.0242 | 1.0 | 3058 | 0.0167 | 0.9967 | 0.9968 | 0.9967 | 0.9967 |
| 0.0146 | 2.0 | 6116 | 0.0163 | 0.9977 | 0.9977 | 0.9977 | 0.9977 |
| 0.009 | 3.0 | 9174 | 0.0112 | 0.9984 | 0.9984 | 0.9984 | 0.9984 |
| 0.0029 | 4.0 | 12232 | 0.0101 | 0.9984 | 0.9984 | 0.9984 | 0.9984 |
| 0.0029 | 5.0 | 15290 | 0.0179 | 0.9980 | 0.9981 | 0.9980 | 0.9980 |
| 0.0012 | 6.0 | 18348 | 0.0160 | 0.9985 | 0.9985 | 0.9985 | 0.9985 |
Framework versions
- Transformers 4.40.2
- Pytorch 2.5.0+cu124
- Datasets 2.18.0
- Tokenizers 0.19.1