Skip to content

fix: correct outdated function reference in docstring #144

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

fix: correct outdated function reference in docstring #144

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
2 changes: 1 addition & 1 deletion flash_mla/flash_mla_interface.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,7 +19,7 @@ def get_mla_metadata(
num_heads_k: The number of k heads.
num_heads_q: The number of q heads. This argument is optional when sparse attention is not enabled
is_fp8_kvcache: Whether the k_cache and v_cache are in fp8 format.
topk: If not None, sparse attention will be enabled, and only tokens in the `indices` array passed to `flash_mla_with_kvcache_sm90` will be attended to.
topk: If not None, sparse attention will be enabled, and only tokens in the `indices` array passed to `flash_mla_with_kvcache` will be attended to.

Returns:
tile_scheduler_metadata: (num_sm_parts, TileSchedulerMetaDataSize), dtype torch.int32.
Expand Down