Skip to content

Add option to disable fast_accumulation for fp8 gemm. #3714

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 1 commit into from

Conversation

jwfromm
Copy link
Contributor

@jwfromm jwfromm commented Feb 19, 2025

Summary: Experimenting with fast accumulation is becoming more relevant as it seems turning it off is needed for FP8 training. This small diff adds the disable_fast_accum flag to our quantize benchmarks to make it easier to evaluate.

Differential Revision: D69863694

@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D69863694

Copy link

netlify bot commented Feb 19, 2025

Deploy Preview for pytorch-fbgemm-docs ready!

Name Link
🔨 Latest commit 62d850c
🔍 Latest deploy log https://app.netlify.com/sites/pytorch-fbgemm-docs/deploys/67b6229cc783d400084817db
😎 Deploy Preview https://deploy-preview-3714--pytorch-fbgemm-docs.netlify.app
📱 Preview on mobile
Toggle QR Code...

QR Code

Use your smartphone camera to open QR code link.

To edit notification comments on pull requests, go to your Netlify site configuration.

Summary:

X-link: facebookresearch/FBGEMM#795

Experimenting with fast accumulation is becoming more relevant as it seems turning it off is needed for FP8 training. This small diff adds the `disable_fast_accum` flag to our quantize benchmarks to make it easier to evaluate.

Differential Revision: D69863694
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D69863694

@facebook-github-bot
Copy link
Contributor

This pull request has been merged in 2fe6f44.

q10 pushed a commit to q10/FBGEMM that referenced this pull request Apr 10, 2025
Summary:
X-link: pytorch#3714

Pull Request resolved: facebookresearch/FBGEMM#795

Experimenting with fast accumulation is becoming more relevant as it seems turning it off is needed for FP8 training. This small diff adds the `disable_fast_accum` flag to our quantize benchmarks to make it easier to evaluate.

Reviewed By: ipiszy, jiawenliu64

Differential Revision: D69863694

fbshipit-source-id: b564113b7fa370bdcf2487b83e6fc91564a1fa09
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants