MoE Expert-Parallel Overlap Skill References - Stable docs: @docs/training/communication-overlap.md - Structured metadata: @skills/nemo-mbridge-perf-expert-parallel-overlap/card.yaml What It Is Expert-parallel (EP) overlap hides the cost of token dispatch/combine all-to-all communication by running it concurrently with expert FFN compute. Optionally, delayed expert weight-gradient computation ( ) provides additional overlap by deferring wgrad to overlap with the next layer's forward. Bridge supports two dispatcher paths: | Dispatcher | Backend | When to use | |---|---|---| | | Standard MoE al…