---
language:
- en
---

This is a large dataset for teaching LLMs to refuse unnmoral,illegal and unethical requests.

This dataset is a mix of:
```
mlabonne/harmful_behaviors 
onepaneai/harmful-prompts-gemini-pro-after-guardrail-evaluation 
datasets/Baidicoot/harmful-rlhf
saddxzcy/Harmful-instruction 
datasets/JailbreakBench/JBB-Behaviors 
Deep1994/ReNeLLM-Jailbreak
LLM-LAT/harmful-dataset 
```