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.

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

Groenenboomj/fixes causal #575

Open
wants to merge 3 commits into
base: main_perf
Choose a base branch
from

Conversation

groenenboomj
Copy link

Add support for causal masking false and some testing changes:
-- backwards kernel support for bf16
-- backwards benchmark

Copy link

@xinyazhang xinyazhang left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@groenenboomj groenenboomj reopened this Aug 12, 2024
@groenenboomj groenenboomj changed the base branch from triton-mlir to main_perf August 12, 2024 16:42
@@ -1259,99 +1261,93 @@ def test_op_varlen_mqa_fwd(Z, HQ, HK, N_CTX, D_HEAD, causal, dtype=torch.float16
#(1, 16, 8192, 63),
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We should test seqlens that are small so 1, 2, 4, 16, 32, 64, 128, 256, etc ...

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants