feat: Enable amp with autocast (fix poor bf16 convergence on GRPO#26
Merged
SahilJain314 merged 2 commits intomainfrom Mar 21, 2025
Merged
feat: Enable amp with autocast (fix poor bf16 convergence on GRPO#26SahilJain314 merged 2 commits intomainfrom
SahilJain314 merged 2 commits intomainfrom
Conversation
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
ashors1
approved these changes
Mar 21, 2025
parthchadha
approved these changes
Mar 21, 2025
KiddoZhu
pushed a commit
that referenced
this pull request
May 6, 2025
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What does this PR do ?
Uses torch autocast to handle mixed precision instead of the FSDP mixed precision object/flag. Seems to work significantly better
purple/pink- new amp
light green - fp32 pure (had convergence issues too due to off-policy)
teal+light pink - old amp, pure bf16
ignore dark green