Move bias add to after the dot operation. #9243
Open
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
copybara-service
bot
force-pushed
the
test_839642272
branch
from
2ea5024 to
1f97479
Compare
Prior functionality added the bias prior to the dot operation by initializing the accumulator along with quantization offsets. This assumes that the bias has the same scale as the inputs. New functionality adds the bias after the dot operation and uses the scale attached to the bias node (not the scale of the inputs). By moving this out, we are able to fuse it with other post-dot scaling operations. PiperOrigin-RevId: 839642272
copybara-service
bot
force-pushed
the
test_839642272
branch
from
1f97479 to
c62249a
Compare
copybara-service
bot
changed the title
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Outdated suggestions cannot be applied.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.
Move bias add to after the dot operation.
Prior functionality added the bias prior to the dot operation by initializing the accumulator along with quantization offsets. This assumes that the bias has the same scale as the inputs.
New functionality adds the bias after the dot operation and uses the scale attached to the bias node (not the scale of the inputs). By moving this out, we are able to fuse it with other post-dot scaling operations.