Skip to content
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

Quantize megablox #1062

Open
wants to merge 10 commits into
base: main
Choose a base branch
from
Open

Quantize megablox #1062

wants to merge 10 commits into from

Conversation

lenscloth
Copy link
Collaborator

Support quantization on megablox

  • Current implementation can accelerate training but not serving, serving acceleration will be done in another PR.
  • Tested correctness after apply quantization on megablox.

@RissyRan
Copy link
Collaborator

We shouldn't add megablox as a copy from JAX repo into MaxText. We are good to work in a dev branch, but in a long term, we need to merge related quantization changes into JAX repo. So MaxText could directly import it.

@lenscloth
Let gmm accept pre-quantized rhs (weight).
fa5d5f3 
By accepting pre-quantized weight, quantized gmm does not need to
quantize weight for every iteration.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@gobbleturk gobbleturk Awaiting requested review from gobbleturk gobbleturk is a code owner

@jonb377 jonb377 Awaiting requested review from jonb377

@khatwanimohit khatwanimohit Awaiting requested review from khatwanimohit khatwanimohit is a code owner

@bvandermoon bvandermoon Awaiting requested review from bvandermoon bvandermoon is a code owner

@vipannalla vipannalla Awaiting requested review from vipannalla vipannalla is a code owner

At least 1 approving review is required to merge this pull request.

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@lenscloth @RissyRan @ZhiyuLi-goog