DeepSpeed
f93e22b3
- Correctness fix PP+ZeRO for gradient accumulation + updates from master (#1263)
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
4 years ago
Correctness fix PP+ZeRO for gradient accumulation + updates from master (#1263) Co-authored-by: Olatunji Ruwase <olruwase@microsoft.com>
References
#1263 - Correctness fix PP+ZeRO for gradient accumulation + updates from master
Author
jeffra
Parents
d6945dea
Loading