Add support for non-power of 2 head size with quantization #2352
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
github-actions
bot
added
the
platform:gpu-intel
petercad
reviewed
Jan 7, 2025
umar456
force-pushed
the
uarshad/sdpa_non_pow2_quan_support
branch
from
a278031 to
9db21cb
Compare
syurkevi
approved these changes
Jan 8, 2025
petercad
reviewed
Jan 8, 2025
petercad
approved these changes
Jan 8, 2025
umar456
force-pushed
the
uarshad/sdpa_non_pow2_quan_support
branch
from
9db21cb to
bdf9a78
Compare
|
make test |
umar456
force-pushed
the
uarshad/sdpa_non_pow2_quan_support
branch
from
bdf9a78 to
563bb62
Compare
umar456
force-pushed
the
uarshad/sdpa_non_pow2_quan_support
branch
from
563bb62 to
50c6cc1
Compare
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Outdated suggestions cannot be applied.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.
Description
This pull request adds the ability to support quantization with head size that is not a power of two. This was a limitation on the vs ugemm and was required by the phi3-mini-4k-instruct model.
The shape that was failing was Q [1x32xSEQ_LENx96] KV [1x32xSEQ_LENx96] which now passes with this change.
Additionally this PR addresses a failure to build when the vs work group was configured with an 16x16 work group tile.