DeepSpeed
Fix zero 1 and 2 CPU-offloaded gradient norm
#7967
Merged

Fix zero 1 and 2 CPU-offloaded gradient norm #7967

delock merged 2 commits into deepspeedai:master from alek6kun:master
alek6kun
alek6kun Fix zero 1 and 2 CPU-offloaded gradient norm
75d7628b
alek6kun alek6kun requested a review from tjruwase tjruwase 60 days ago
alek6kun alek6kun requested a review from tohtana tohtana 60 days ago
chatgpt-codex-connector
delock
alek6kun
delock
delock commented on 2026-04-10
delock
alek6kun Add regression test
94b674c7
alek6kun alek6kun requested a review from loadams loadams 60 days ago
alek6kun
delock
delock approved these changes on 2026-04-11
delock delock merged 9d632f10 into master 60 days ago

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone