Skip to content

SFT: device-agnostic chunked linear+CE loss and --gradient-checkpointing forwarding - #330

Merged
michaellchung merged 2 commits into
mainfrom
sft-chunked-ce
Nov 3, 2026
Merged

SFT: device-agnostic chunked linear+CE loss and --gradient-checkpointing forwarding#330
michaellchung merged 2 commits into
mainfrom
sft-chunked-ce

Forward --gradient-checkpointing from the launcher to torch learners

128e1ef
Select commit
Loading
Failed to load commit list.

Select a check to view from the sidebar