DeepSpeed
Preserve convolution shapes and scaling in Muon updates
#8552
Merged

Preserve convolution shapes and scaling in Muon updates #8552

vineethsaivs
vineethsaivs vineethsaivs requested a review from tjruwase tjruwase 64 days ago
vineethsaivs vineethsaivs requested a review from tohtana tohtana 64 days ago
vineethsaivs vineethsaivs requested a review from loadams loadams 64 days ago
vineethsaivs
pengdurice
pengdurice commented on 2026-09-16
delock delock requested a review from delock delock 63 days ago
delock delock removed review request from delock delock 62 days ago
delock delock removed review request from tjruwase tjruwase 62 days ago
delock delock removed review request from tohtana tohtana 62 days ago
delock delock removed review request from loadams loadams 62 days ago
delock delock requested a review from pengdurice pengdurice 62 days ago
pengdurice
vineethsaivs
alanhuangyoo
alanhuangyoo commented on 2026-09-23
alanhuangyoo
pengdurice pengdurice enabled auto-merge 57 days ago
disabled auto-merge 57 days ago
Manually disabled by user
pengdurice
pengdurice approved these changes on 2026-09-23
pengdurice pengdurice enabled auto-merge 57 days ago
pengdurice
alanhuangyoo
pengdurice
disabled auto-merge 57 days ago
Head branch was pushed to by a user without write access
vineethsaivs
alanhuangyoo
pengdurice
vineethsaivs Preserve convolution shapes and scaling in Muon updates
456e64f2
vineethsaivs vineethsaivs force pushed from f7fbef5b to 456e64f2 55 days ago
vineethsaivs
hwchen2017 hwchen2017 requested a review from copilot-pull-request-reviewer copilot-pull-request-reviewer 51 days ago
copilot-pull-request-reviewer
copilot-pull-request-reviewer commented on 2026-09-29
gpu-test-labeler gpu-test-labeler added skip-gpu-test
vineethsaivs
pengdurice Merge branch 'master' into fix-muon-convolution-shape-20260916
604fd345
gpu-test-labeler gpu-test-labeler removed skip-gpu-test
gpu-test-labeler gpu-test-labeler added skip-gpu-test
pengdurice pengdurice enabled auto-merge 21 hours ago
pengdurice pengdurice merged 64616608 into master 20 hours ago

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone