Megatron-DeepSpeed
ab0a7f8f
- Loading universal checkpoint with reshaping
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
3 years ago
Loading universal checkpoint with reshaping
References
#239 - Reshape deepspeed checkpoint
#289 - No-ZeRO reshaping
#292 - a branch combining layer-norm-auto-sync and ds_ckpt_reshape
Author
tjruwase
Parents
d3005120
Loading