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

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

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

e35cb28
Select commit
Loading
Failed to load commit list.
Google CLA / cla/google succeeded Oct 19, 2026 in 8s

✅ All contributors are covered under a CLA with Google

See https://cla.developers.google.com/ for more info about Google's Contributor License Agreement (CLA).

ℹ️ Googlers: Go here to view more details and manage scans for this pull request.

Details

The following contributors were found for this pull request:

✅ e35cb28 Author: @AusarYao28 <aus*****00​@gmail.com>, <ya*****en​@google.com>

(Only the first commit for a unique contributor is listed.)