Datasets:

LucidityAI
/

LLM-Alignment

Modalities:

Formats:

Languages:

Size:

Libraries:

License:

LLM-Alignment / README.md

Create README.md

b54e6b2 verified 18 days ago

266 Bytes

metadata

license: apache-2.0
language:
  - en

LLM Alignment V1

This dataset contains jailbreaks from TrustAIRLab/in-the-wild-jailbreak-prompts and JailbreakV-28K/JailBreakV-28k.

The dataset includes a pre-defined refusal message for both thinking and non-thinking.