NVIDIA / NVIDIA/Megatron-LM

[QUESTION] Purpose of all2all communication for probs in MoEAlltoAllTokenDispatcher

Open
#1,716 3 comments 0 reactions 1 assignee Claimed by @yanring View on GitHub
community-request module: moe question
Dominant language
Python
Stars
17.9k
Forks
4.5k
Avg merge
4d 6h
Merged PRs (30d)
271

Description

In the `ADLR/megatron-lm!2668 - perf(MoE): Memory efficient token permutation`, I noticed an all2all communication is applied to the probs tensor in `MoEAlltoAllTokenDispatcher` . Could someone tell me the motivation for doing this? I didn't observe a drop in memory usage compared to previous versions.

Appreciate any insights or references to discussions/design documents!

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.