English
Safety
Defense
Jailbreak
Multi-turn
Harmful
Benign

Improve model card metadata and add usage instructions

#1
by nielsr HF Staff - opened

Hi! I'm Niels from the Hugging Face community science team.

This PR improves the model card for TurnGate-0.1 by:

  • Adding the pipeline_tag: text-classification to the YAML metadata.
  • Including Quick Start instructions for evaluating baselines and trained checkpoints, based on the GitHub README.
  • Adding details about the Online Battle (adversarial evaluation) and the MTID Dataset structure.
  • Linking to the research paper in the Markdown section.

These changes make the repository more accessible to researchers looking to reproduce or build upon your work!

Ready to merge
This branch is ready to get merged automatically.

Sign up or log in to comment