DeepSpeed
Fix AutoEP global L2 clipping for sharded expert ownership
#8476
Open

Fix AutoEP global L2 clipping for sharded expert ownership #8476

gss10282023
gss10282023 Fix AutoEP global L2 clipping for sharded expert ownership
8288645c
gss10282023 Merge branch 'master' into fix-autoep-ownership-aware-clipping
e05fa3fa
gss10282023 gss10282023 marked this pull request as ready for review 18 hours ago
gss10282023 gss10282023 requested a review from tjruwase tjruwase 18 hours ago
gss10282023 gss10282023 requested a review from tohtana tohtana 18 hours ago
gss10282023 gss10282023 requested a review from hwchen2017 hwchen2017 18 hours ago
gss10282023 gss10282023 requested a review from loadams loadams 18 hours ago
chatgpt-codex-connector
chatgpt-codex-connector commented on 2026-09-10
chatgpt-codex-connector
gss10282023 Fix AutoEP clipping across mesh sequence-parallel ranks
d243a807

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone