Added softmax to autocast policy #8509
Merged
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
|
Hi @avizon-aws, thank you for sending the PR! Could you force push again to trigger TPU CI? |
lsy323
approved these changes
Dec 20, 2024
|
@avizon-aws LGTM, thanks! |
avizon-aws
added a commit
that referenced
this pull request
Dec 20, 2024
This was referenced Dec 20, 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Outdated suggestions cannot be applied.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.
Softmax operation was not added to the autocast policy previously which led to its computation in BF16. Since this is a non linear operation, we want to preserve precision and perform it in FP32. This PR adds its support as well as a HLO based unit test.