In the flash attention RMSNorm and positional encoding pass, hoist the work that - #1040
Merged
Merged
Conversation
copybara-service
Bot
force-pushed
the
test_979182129
branch
2 times, most recently
from
September 25, 2026 13:19
e05638c to
91f473e
Compare
…e work that does not depend on the attention head out of the per-head loop: the token position and the row index are now computed once per token. Filter `--gtest_*` flags in `gemma_test` so it can run under a test runner that passes them. PiperOrigin-RevId: 988257504
copybara-service
Bot
force-pushed
the
test_979182129
branch
from
September 25, 2026 14:07
91f473e to
3806ff7
Compare
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
In the flash attention RMSNorm and positional encoding pass, hoist the work that
does not depend on the attention head out of the per-head loop: the token
position and the row index are now computed once per token.
Filter
--gtest_*flags ingemma_testso it can run under a test runner thatpasses them.