version update

Bump version to 0.9.0

removed .dev0 from version.py

commit change

v_bump

bump version

Updt triton pin (#89)

* Update setup.py

make torch pin more flexible

* Update setup.py

Merge pull request #47 from stanford-futuredata/fix-topology-kernel

Fix bug in topology kernel for ffn_hidden_size>4096.

Merge pull request #34 from stanford-futuredata/fsdp_refactor

Refactoring class hierarchy for FSDP wrapping

Merge pull request #31 from vchiley/no_bias

Enable running MegaBlocks MoE without bias