|
|
--- |
|
|
license: apache-2.0 |
|
|
datasets: |
|
|
- lmsys/toxic-chat |
|
|
- PKU-Alignment/BeaverTails |
|
|
- lmsys/lmsys-chat-1m |
|
|
language: |
|
|
- en |
|
|
metrics: |
|
|
- f1 |
|
|
- accuracy |
|
|
tags: |
|
|
- ai-safety |
|
|
- safetyguard |
|
|
- safety |
|
|
- benchmark |
|
|
- mistral |
|
|
- salad-bench |
|
|
- evluation |
|
|
--- |
|
|
# MD-Judge for Salad-Bench |
|
|
|
|
|
|
|
|
## Model Details |
|
|
|
|
|
MD-Judge is a LLM-based safetyguard, fine-tund on top of [Mistral-7B](https://huggingface.co/mistralai/Mistral-7B-v0.1). MD-Judge serves as a classifier to evaluate the safety of QA pairs. |
|
|
|
|
|
MD-Judge was born to study the safety of different LLMs serving as an general evaluation tool, which is proposed under the [SALAD-Bench paper]() |
|
|
|
|
|
- **Developed by:** The SALAD-Bench Team |
|
|
- **Model type:** An auto-regressive language model based on the transformer architecture. |
|
|
|
|
|
## Model Sources |
|
|
|
|
|
- **Repository:** [SALAD-Bench Github](https://github.com/OpenSafetyLab/SALAD-BENCH) |
|
|
- **Dataset:** Coming soon |
|
|
- **Paper:** Coming soon |
|
|
|
|
|
## Uses |
|
|
Please refer to our [Github](https://github.com/OpenSafetyLab/SALAD-BENCH) for more using examples |
|
|
|
|
|
```python |
|
|
|
|
|
``` |
|
|
|
|
|
## Citation |
|
|
|
|
|
**BibTeX:** |
|
|
|
|
|
|
