OpenMOSE

/

HRWKV7-hxa079-Qwen3-8B

like 0

This PR enhances the model card for HRWKV7-hxa079-Qwen3-8B by:

• Adding Metadata: Included pipeline_tag: text-generation to ensure better discoverability on the Hugging Face Hub, and library_name: transformers to enable the "how to use" widget, as the model is compatible with the Transformers library via trust_remote_code=True.
• Prominent Links: Added direct links to the associated paper (RADLADS: Rapid Attention Distillation to Linear Attention Decoders at Scale) and the main GitHub repository (https://github.com/recursal/RADLADS) at the top of the card. The existing code links within the "Thank you" and "Training Code" sections have been clarified.
• Sample Usage: Provided a correct transformers-based code snippet for text generation, guiding users on how to run inference with the model.
• Citation: Added the BibTeX citation from the paper's repository.

These changes improve the model's visibility, usability, and provide more comprehensive information for users and researchers.