LLM jailbreak + safety data for chatbot apps
A dataset for training and evaluating LLMs to detect jailbreakable and benign
What you can do with it
Study or compare language-model prompts and responses.
Download size: 666.9 KB
Resource information
- Data purpose
- Training data
- Task
- Language models and prompts
- Modality
- Text
- Version
- 1
- Updated by publisher
- Jan 15, 2025
- License
- Apache 2.0