Add option to replicate attn weights on FSDP or EP#3480
Open
gobbleturk wants to merge 7 commits intomainfrom
Open
Add option to replicate attn weights on FSDP or EP#3480gobbleturk wants to merge 7 commits intomainfrom
gobbleturk wants to merge 7 commits intomainfrom
Conversation
gobbleturk
requested review from
A9isha,
NicoGrande,
NuojCheng,
RissyRan,
SurbhiJainUSC,
aireenmei,
bvandermoon,
dipannita08,
gagika,
hengtaoguo,
igorts-git,
jesselu-google,
jiangjy1982,
khatwanimohit,
richjames0,
shralex,
suexu1025 and
vipannalla
as code owners
Codecov Report✅ All modified and coverable lines are covered by tests. 📢 Thoughts on this report? Let us know! |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Not meant to be submitted.
Gives an easy (CLI) option for how to shard embed (embed-attn) one of three ways:
We give these options and set only EP as default since it is more performant for large scale moe runs - we don't want to shard the embed dimension 4096 ways, but only EP=64 ways (we use EP since EP is 2D in our best configs)