Megatron-DeepSpeed
a6b26240
- Move norm to batch pipe
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
3 years ago
Move norm to batch pipe
References
#326 - Add option to normalize loss per target
#336 - Add multiple evaluation compat
Author
Muennighoff
Parents
2e7554d7
Loading