WNA16 does not apply optimized RTN for moe layers by default #1245
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
for more information, see https://pre-commit.ci
Signed-off-by: Zhang, Weiwei1 <[email protected]>
for more information, see https://pre-commit.ci
…ix_0108 # Conflicts: # auto_round/compressors/base.py
for more information, see https://pre-commit.ci
wenhuach21
changed the title
for more information, see https://pre-commit.ci
Contributor
There was a problem hiding this comment.
Pull request overview
This PR addresses an issue where optimized RTN (Round-to-Nearest) quantization was not properly disabled by default for MoE (Mixture of Experts) layers in WNA16 configurations. The change introduces MoE model detection and automatically disables optimized RTN for expert layers to improve efficiency, while allowing users to override this behavior with --enable_opt_rtn.
Key Changes:
- Added
is_moe_model()utility function to detect MoE models by examining config keys and module names - Changed default value of
disable_opt_rtnfromTruetoNoneto enable automatic optimization detection - Implemented logic to automatically disable optimized RTN for MoE expert layers unless explicitly enabled by the user
Reviewed changes
Copilot reviewed 3 out of 3 changed files in this pull request and generated 2 comments.
| File | Description |
|---|---|
| auto_round/utils/model.py | Adds is_moe_model() function to detect MoE models through config inspection and module name checking |
| auto_round/compressors/config.py | Changes disable_opt_rtn default from True to None to allow automatic optimization |
| auto_round/compressors/base.py | Implements MoE detection and automatic optimized RTN disabling for expert layers, with improved logging and user override support |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
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.
No description provided.