Augment blockwise quantization #18101
Merged
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-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
yufenglee
reviewed
Oct 25, 2023
yufenglee
reviewed
Oct 25, 2023
yufenglee
reviewed
Oct 26, 2023
edgchen1
reviewed
Oct 27, 2023
edgchen1
reviewed
Oct 27, 2023
edgchen1
previously approved these changes
Oct 27, 2023
yufenglee
approved these changes
Oct 30, 2023
chenfucn
added a commit
that referenced
this pull request
Nov 3, 2023
### Description Replace block-wise 4b quantization implementation ### Motivation and Context In #18101 we have an augmented block-wise 4b quantization interface and implementation. Here we use this new implementation in onnxruntime contrib ops --------- Co-authored-by: Edward Chen <[email protected]>
kleiti
pushed a commit
to kleiti/onnxruntime
that referenced
this pull request
Mar 22, 2024
### Description Augment block wise 4b quantization -- plain CPU impl ### Motivation and Context Allow column wise or row wise blocks. Experiments show row wise quantization in LLM weight matrices achieves better precision. Added tests for quantization and dequantization code.
kleiti
pushed a commit
to kleiti/onnxruntime
that referenced
this pull request
Mar 22, 2024
### Description Replace block-wise 4b quantization implementation ### Motivation and Context In microsoft#18101 we have an augmented block-wise 4b quantization interface and implementation. Here we use this new implementation in onnxruntime contrib ops --------- Co-authored-by: Edward Chen <[email protected]>
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Outdated suggestions cannot be applied.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
Augment block wise 4b quantization -- plain CPU impl
Motivation and Context
Allow column wise or row wise blocks. Experiments show row wise quantization in LLM weight matrices achieves better precision.
Tests for quantization and dequantization code.