DeepSpeed
d26c258b
- added shaden's set_train_batch_size patches, plus formatting
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
4 years ago
added shaden's set_train_batch_size patches, plus formatting
References
megatron2.4-3d
#1400 - Big science fix passing multiple tensors
Author
jeffra
Parents
081ddb5f
Loading