[WIP] Add non-gated colvec_scale support to gemm_dact#143
Draft
JungHoyoun wants to merge 2 commits into
Draft
Conversation
JungHoyoun
changed the title
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
I would like to add a
colvec_scaleargument to thegemm_dactsignature, similar togemm_dgated.The motivation is to add
squared_reluactivation support tosonic-moe. Currently,sonic-moeonly supports GLU-family activations, and this change is intended to extend support to architectures such as Nemotron 3.The initial commit was written with the help of Codex. I will clean up and verify the code properly before moving this PR out of draft.
So far, testing has only been run on an SM90 environment.
Also, #142 was accidentally pushed while I was working with an AI tool. I have closed that PR, but I would appreciate it if it could be deleted from the repository history if possible. Sorry for the noise.