Skip to content

Use the sequence-wise balance loss for DeepSeek-V4 routers - #5850

Open
AusarYao28 wants to merge 1 commit into
mainfrom
yaoyuchen/ds4-seq-aux-sqrtsoftplus
Open

Use the sequence-wise balance loss for DeepSeek-V4 routers#5850
AusarYao28 wants to merge 1 commit into
mainfrom
yaoyuchen/ds4-seq-aux-sqrtsoftplus

Commits

  1. Commits on Oct 20, 2026