Currently multi-gpu generate does not work with hf.generate for hf checkpoints. This PR fixes that. #1332
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
vchiley
reviewed
Jul 2, 2024
vchiley
reviewed
Jul 2, 2024
snarayan21
reviewed
Jul 2, 2024
vchiley
approved these changes
Jul 2, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Multi gpu generation using hf.generate with device map = 'auto' does pipeline parallelism and moves different modules to different gpus. This results in input tensors to certain operations being on different gpus than other inputs to that operation, which results in an error. This PR moves the tensors to match the other tensors. This should not slow down training because during training all of these tensor movements should be no-ops.