main
[fp8] Disable all_gather intranode. Disable Redundant all_gather fp8 (#6059)
[misc] fit torch api upgradation and remove legecy import (#6093)
[shardformer]: support gpt-j, falcon, Mistral and add interleaved pipeline for bert (#5088)
[checkpointio]support asyncio for 3d (#6152)
[Device]Support npu (#6159)
[chore] remove unnecessary assert since compute list might not be recorded
[npu] change device to accelerator api (#5239)