husj576

/

GRIFFIN-llama2-chat-13B

like 0

This PR significantly enhances the model card for the GRIFFIN model by:

• Adding pipeline_tag: text-generation to the metadata for better discoverability on the Hub.
• Specifying library_name: transformers to enable the automated "How to use" widget, as evidenced by the config.json and the provided usage code.
• Linking to the official paper: GRIFFIN: Effective Token Alignment for Faster Speculative Decoding.
• Providing a link to the GitHub repository: https://github.com/hsj576/GRIFFIN.
• Including a detailed "Overview" section and relevant benchmark images and an acceleration GIF from the GitHub README.
• Adding a sample usage code snippet, directly extracted and slightly improved from the GitHub README, to demonstrate how to run inference.
• Including the BibTeX citation.

These additions will make the model more informative, discoverable, and easier to use for the community.