Skip to content

[grug] Multi-host MoE training fails with ShardingTypeError on shared-expert residual add #4312

[grug] Multi-host MoE training fails with ShardingTypeError on shared-expert residual add

[grug] Multi-host MoE training fails with ShardingTypeError on shared-expert residual add #4312

Triggered via issue June 9, 2026 04:59
@rjpowerrjpower
commented on #6296 1c9b281
Status Skipped
Total duration 1s
Artifacts

ops-claude.yaml

on: issue_comment
Fit to window
Zoom out
Zoom in