1 dataset found

Tags: Hurtfulness

Filter Results
  • HONEST

    HONEST is a fairness dataset specifically designed to assess LMs' outputs' hurtfulness.
You can also access this registry using the API (see API Docs).