-
Notifications
You must be signed in to change notification settings - Fork 284
Fix Per Tensor 3d rehsape #2293
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
Conversation
stack-info: PR: #2293, branch: drisspg/stack/64
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/2293
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit a18d989 with merge base 9d5b9ad ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
stack-info: PR: #2293, branch: drisspg/stack/64
stack-info: PR: #2293, branch: drisspg/stack/64
Pulling this down to try it out on Flux.. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Works for me, thanks for the fix :)
Ideally, as we discussed offline, determination of rowwise scaling should be fixed in torch as well: https://github.com/pytorch/pytorch/blob/main/torch/_inductor/kernel/mm_common.py#L122
return output + bias | ||
return torch._scaled_mm( |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
trimmin down those lines!
@jbschlosser you wanna add yourself as a contributor so I can land this? |
@drisspg oh yeah, good call |
@drisspg try merging now? |
@jbschlosser magic |
Stacked PRs:
Fix Per Tensor 3d rehsape