Improve model card: update license, add paper details, usage, results, and citation for HAPO Qwen2.5-Math-7B

#1
by nielsr HF Staff - opened

This PR significantly enhances the model card for the Qwen2.5-Math-7B-32k-HAPO-Base model by:

  • Updating the license: Changed from mit to apache-2.0 to reflect the official license found in the GitHub repository.
  • Adding comprehensive paper details: Including the paper's full title, a direct link to the Hugging Face paper page, and the complete abstract.
  • Integrating GitHub content: Incorporating the "About", "Installation", "Usage" (Preparation, Training, Evaluation), "Results", and "Training Dynamics" sections, complete with relevant figures from the GitHub README.
  • Providing a GitHub repository link: Making it easy for users to find the full code.
  • Completing the BibTeX citation: Adding the full and accurate BibTeX entry for the paper.
  • Adding contact information and acknowledgements: From the GitHub README for community engagement.

These updates aim to provide a much richer and more accurate overview of the model, ensuring users have all necessary information to understand, implement, and cite HAPO.

Cannot merge
This branch has merge conflicts in the following files:
  • README.md

Sign up or log in to comment