Skip to content

Remove obsolete Tokamax gmm pipeline FSDP workaround #4214

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
huytransformer wants to merge 1 commit into
base: main
Choose a base branch
from
+1 −4
Open

Remove obsolete Tokamax gmm pipeline FSDP workaround #4214

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 1 addition & 4 deletions src/maxtext/layers/moe.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1247,10 +1247,7 @@ def jax_ragged_dot_gmm(inputs, kernel, tiling, group_sizes, expert_assignments,
return output

def get_tokamax_group_sizes(group_sizes, inputs, kernel):
# TODO (b/491979205) pipeline fsdp ag per repeat fails tokamax gmm
if self.config.use_qwix_quantization or (
self.config.using_pipeline_parallelism and self.config.pipeline_fsdp_ag_per_repeat
):
if self.config.use_qwix_quantization:
return group_sizes
elif self.config.attention == "vllm_rpa":
return group_sizes
Expand Down
Loading