Cast bf16 to fp32 for sum/reduce #6974
Closed
+4
−2
Conversation
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
|
I believe the change was on purpose, the description was unfortunately outdated, see this discussion: In any case I think the current behavior is what we want, we should only upcast if user explicitly pass float32 as accumulator type |
|
closing this for now based on comments, feel free to start a new discussion if you think this is still needed |
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Outdated suggestions cannot be applied.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.
There used to be this behavior but got stopped by f9ad25e#diff-7935be9afb58d190ea8fd7a849bdc375fa48f75aff33c7bb152428feaa6860f4 by @Mogball
By reading the PR summary, it seems unintentional to stop promoting bf16 to fp32.