Skip to content
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

misc #44

Open
wants to merge 1 commit into
base: drisspg/stack/6
Choose a base branch
from
Open

misc #44

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
8 changes: 4 additions & 4 deletions transformer_nuggets/flex/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,16 +10,16 @@ class FlexAttentionKernelArgs:

# Performance tuning options
num_warps: Optional[int] = None
"""Number of warps to use in the CUDA kernel. If None, will be autotuned."""
"""Number of warps to use in the CUDA kernel."""

num_stages: Optional[int] = None
"""Number of pipeline stages to use in the CUDA kernel. If None, will be autotuned."""
"""Number of pipeline stages to use in the CUDA kernel."""

BLOCK_M: Optional[int] = None
"""Thread block size across the seqlen dim of Q. If None, will be autotuned."""
"""Thread block size across the seqlen dim of Q."""

BLOCK_N: Optional[int] = None
"""Block size to iterate over across the seqlen dim of K/V in each thread block. If None, will be autotuned."""
"""Block size to iterate over across the seqlen dim of K/V in each thread block."""

# Numerical behavior options
PRESCALE_QK: bool = False
Expand Down
Loading