Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

refactor dbrx experts to use FusedMoe layer #186

Merged
merged 5 commits into from
Sep 17, 2024
Merged

Conversation

divakar-amd
Copy link

@divakar-amd divakar-amd commented Sep 16, 2024

Re-use FusedMoe layer for dbrx model.

  • Create DbrxMoe class
  • DbrxExperts class now uses FusedMoe as the base class
    • with custom weight loader
  • This will allow us to use existing moe based optimization from mixtral like VLLM_MOE_PADDING
  • Will also expose the "process_weights_after_loading" for this model

@ROCm ROCm deleted a comment from github-actions bot Sep 16, 2024
@divakar-amd divakar-amd marked this pull request as ready for review September 16, 2024 19:22
@charlifu
Copy link

LGTM, thank you for making the change.

@divakar-amd divakar-amd merged commit 6bd99d2 into main Sep 17, 2024
16 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants