bump to 0.3.16 

bump to 0.3.15 

bump to 0.3.14 

bump version 0.3.13 

bump to v0.3.12 

bump to 0.3.11 

version bump to 0.3.10 

Elastic training support (deepspeedai#602) Co-authored-by: Samyam Rajbhandari <samyamr@microsoft.com>

bump to 0.3.8 

calculate grad norm wrt sub partitions