transformers
fa83baba
- Fix Doge load balancing loss pairing expert loads with rank-ordered p… (#50266)
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
1 day ago
Fix Doge load balancing loss pairing expert loads with rank-ordered p… (#50266) Fix Doge load balancing loss pairing expert loads with rank-ordered probabilities
References
#50266 - Fix Doge load balancing loss pairing expert loads with rank-ordered p…
Author
Andreas-Hum
Parents
89e9737d
Loading