update rotation doc.#1999
Merged
Merged
Conversation
Signed-off-by: lkk12014402 <kaokao.lv@intel.com>
lkk12014402
requested review from
wenhuach21 and
xin3he
and removed request for
xin3he
for more information, see https://pre-commit.ci
Contributor
|
/azp run Unit-Test-CUDA-AutoRound |
|
Azure Pipelines could not run because the pipeline triggers exclude this branch/path. |
chensuyue
approved these changes
Jul 6, 2026
Contributor
There was a problem hiding this comment.
Pull request overview
Note
Copilot couldn't run its full agentic review because no GitHub Actions runner was available. Make sure your repository has a runner available to run Copilot's review, or add a copilot-setup-steps.yml file specifying one with the runs-on attribute. See the docs for more details.
Simplifies the step-by-step rotation documentation in both English and Chinese, and adds a dedicated “Rotation Details” reference including example accuracy results.
Changes:
- Condensed the Rotation (Experimental) sections in
step_by_step.mdandstep_by_step_CN.mdto a short API-focused summary. - Added
docs/rotation_details.mdas an in-depth rotation reference, including configuration details and lm-eval results tables.
Reviewed changes
Copilot reviewed 3 out of 3 changed files in this pull request and generated 2 comments.
| File | Description |
|---|---|
| docs/step_by_step.md | Replaces the long rotation explanation with a brief usage snippet and a link to the new detailed doc. |
| docs/step_by_step_CN.md | Mirrors the English simplification and links to the new detailed doc (English). |
| docs/rotation_details.md | New detailed rotation reference covering modes/configs, dispatch behavior, and example accuracy results. |
| Rotation redistributes outliers in weights and activations before quantization, making the distribution more uniform and quantization-friendly. It is most useful for aggressive low-bit schemes such as MXFP4, NVFP4 and W4A4. | ||
|
|
||
| Two rotation approaches are available: | ||
| AutoRound applies rotation through the `rotation_config` argument. The `"quarot"` preset — deterministic Hadamard rotation (QuaRot / SpinQuant), no training and no calibration data — is recommended for most use cases. |
| 旋转在量化前对权重和激活中的离群点进行重分布，使分布更加均匀、对量化更友好。它对 MXFP4、NVFP4、W4A4 等激进的低比特方案最为有效。 | ||
|
|
||
| 目前提供两种旋转方式： | ||
| AutoRound 通过 `rotation_config` 参数应用旋转。推荐在大多数场景中使用 `"quarot"` 预设——确定性 Hadamard 旋转（QuaRot / SpinQuant），无需训练、无需校准数据。 |
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
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
step_by_step*documents