DeepSpeed
Sum GPTBigCode's shared key/value gradient across tensor-parallel ranks
#8928

Merged

Sum GPTBigCode's shared key/value gradient across tensor-parallel ranks #8928

delock merged 1 commit into deepspeedai:master from Achyuthan-S:fix-gptbigcode-mqa-kv-grad

Sum GPTBigCode's shared key/value gradient across tensor-parallel ranks

0bdd7b73

Achyuthan-S requested a review from

copilot-pull-request-reviewer 8 days ago

Achyuthan-S requested a review from

hwchen2017 8 days ago

Achyuthan-S requested a review from

loadams 8 days ago

Achyuthan-S requested a review from

tjruwase 8 days ago

Achyuthan-S requested a review from

tohtana 8 days ago

copilot-pull-request-reviewer commented on 2026-10-27

chatgpt-codex-connector commented on 2026-10-27

delock requested a review from

delock 7 days ago

delock approved these changes on 2026-10-29

delock merged 9148ab22 into master 5 days ago

Achyuthan-S deleted the fix-gptbigcode-mqa-kv-grad branch 5 days ago

Reviewers

delock

hwchen2017

loadams

tjruwase

tohtana

Assignees

No one assigned

Labels

None yet

Milestone

No milestone