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

Adding "longrope" for Phi-3 (#2172) #2179

Merged
merged 1 commit into from
Jul 5, 2024
Merged

Conversation

amihalik
Copy link
Contributor

@amihalik amihalik commented Jul 3, 2024

What does this PR do?

Updates the rope scaling type check for Phi-3 to include "longrope".

I implemented and tested these changes using this process:

  1. Launched latest TGI docker container on an aws g6.2xlarge
  2. Modified /opt/conda/lib/python3.10/site-packages/text_generation_server/layers/rotary.py in the container
  3. Ran text-generation-launcher --model-id microsoft/Phi-3-mini-128k-instruct --trust-remote-code
  4. Validated that Phi-3 loaded
  5. Executed queries against the endpoint and validated responses from TGI

Fixes #2172

Before submitting

  • This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
  • Did you read the contributor guideline,
    Pull Request section?
  • Was this discussed/approved via a Github issue or the forum? Please add a link
    to it if that's the case.
  • Did you make sure to update the documentation with your changes? Here are the
    documentation guidelines, and
    here are tips on formatting docstrings.
  • Did you write any new necessary tests?

Who can review?

Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.

Copy link
Member

@danieldk danieldk left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Awesome, from a diff against their two model revisions it indeed seems that they just renamed the embeddings from su to longrope.

Thanks!

@danieldk danieldk merged commit c6bcadf into huggingface:main Jul 5, 2024
ErikKaum pushed a commit that referenced this pull request Jul 26, 2024
yuanwu2017 pushed a commit to yuanwu2017/tgi-gaudi that referenced this pull request Sep 26, 2024
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.

Phi-3's new rope scaling type longrope not supported
2 participants