Skip to content

Megatron-Bridge distillation: packed chat-SFT data prep and per-document attention - #2912

Merged
kevalmorabia97 merged 6 commits into
mainfrom
kmorabia/sft-packing-upstream
Oct 23, 2026
Merged

kevalmorabia97 merged 6 commits into
mainfrom
kmorabia/sft-packing-upstream

Keep assistant-less rows for loss_mode full; raise chat-tokenization …

17964ed
Select commit
Loading
Failed to load commit list.
DCO / DCO succeeded Oct 23, 2026 in 0s

DCO

All commits are signed off!