Skip to content

Use a symmetric quantization with no clipping error to improve llama perplexity #5163

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.

Sign up for GitHub

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

Jump to bottom
Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Use a symmetric quantization with no clipping error to improve llama perplexity #5163

wants to merge 1 commit into from

Conversation

iseeyuan
Copy link
Contributor

@iseeyuan iseeyuan commented Sep 7, 2024

Refer to pytorch/ao#805 for the details.
With this change, the perplexity of a llama model is improved 4% on wikitext.

@pytorch-bot PyTorch Bot
Copy link

pytorch-bot bot commented Sep 7, 2024 •
edited
Loading

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/5163

Note: Links to docs will display an error until the docs builds have been completed.

✅ No Failures

As of commit 04e2465 with merge base a4be79f (image):
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

@facebook-github-bot facebook-github-bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Sep 7, 2024
@facebook-github-bot
Copy link
Contributor

@iseeyuan has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.

facebook-github-bot pushed a commit that referenced this pull request Sep 7, 2024
@iseeyuan @facebook-github-bot
Use a symmetric quantization with no clipping error to improve llama …
6de5818 
…perplexity (#5163)

Summary:
Refer to pytorch/ao#805 for the details.
With this change, the perplexity of a llama model is improved 4% on wikitext.


Differential Revision: D62342523

Pulled By: iseeyuan
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D62342523

facebook-github-bot pushed a commit that referenced this pull request Sep 9, 2024
@iseeyuan @facebook-github-bot
Use a symmetric quantization with no clipping error to improve llama …
129d673 
…perplexity (#5163)

Summary:
Refer to pytorch/ao#805 for the details.
With this change, the perplexity of a llama model is improved 4% on wikitext.


Differential Revision: D62342523

Pulled By: iseeyuan
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D62342523

facebook-github-bot pushed a commit that referenced this pull request Sep 11, 2024
@iseeyuan @facebook-github-bot
Use a symmetric quantization with no clipping error to improve llama …
2db6f07 
…perplexity (#5163)

Summary:
Refer to pytorch/ao#805 for the details.
With this change, the perplexity of a llama model is improved 4% on wikitext.


Reviewed By: mergennachin, helunwencser

Differential Revision: D62342523

Pulled By: iseeyuan
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D62342523

@iseeyuan @facebook-github-bot
Use a symmetric quantization with no clipping error to improve llama …
04e2465 
…perplexity (#5163)

Summary:
Refer to pytorch/ao#805 for the details.
With this change, the perplexity of a llama model is improved 4% on wikitext.


Reviewed By: mergennachin, helunwencser

Differential Revision: D62342523

Pulled By: iseeyuan
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D62342523

@github-actions GitHub Actions
Copy link

Looks like this PR hasn't been updated in a while so we're going to go ahead and mark this as Stale.
Feel free to remove the Stale label if you feel this was a mistake.
If you are unable to remove the Stale label please contact a maintainer in order to do so.
If you want the bot to never mark this PR stale again, add the no-stale label.
Stale pull requests will automatically be closed after 30 days of inactivity.

@github-actions github-actions bot added the stale PRs inactive for over 60 days label Aug 27, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@mergennachin mergennachin mergennachin approved these changes

@helunwencser helunwencser helunwencser approved these changes

Assignees

No one assigned

Labels

CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. fb-exported stale PRs inactive for over 60 days

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

4 participants

@iseeyuan @facebook-github-bot @mergennachin @helunwencser