-
Notifications
You must be signed in to change notification settings - Fork 38
[Bug fix] Fix log probabilities handling #1114
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
Conversation
DescriptionStart with a short description of what the PR does and how this is a change from The rest of the description includes relevant details and context, examples:
If the change fixes a bug or a Github issue, please include a link, e.g.,: TestsPlease describe how you tested this change, and include any instructions and/or ChecklistBefore submitting this PR, please make sure:
|
vipannalla
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
looks good
3a2aa5a to
6258617
Compare
Description
Upstream PR (vllm-project/vllm#28245) broke our logprob handling, and this PR fixes this issue.
Instead of returning a list of floating points as log prob, upstream now expects a numpy array.
This PR also fixes a minor padding issue that was introduced by a recent tpu_inference PR.
Tests
Updated the e2e DP test.
https://buildkite.com/tpu-commons/tpu-inference-ci/builds/5646
Checklist
Before submitting this PR, please make sure: