Skip to content

Deleted activation24, fp8 gemm with 2:4 sparsity#3744

Merged
howardzhang-cv merged 1 commit into
mainfrom
BE_task/delete_activation24_again
Jan 28, 2026
Merged

Deleted activation24, fp8 gemm with 2:4 sparsity#3744
howardzhang-cv merged 1 commit into
mainfrom
BE_task/delete_activation24_again

Conversation

@howardzhang-cv

Copy link
Copy Markdown
Contributor

Summary: Deleted torchao/csrc/cuda/activation24
Included sparsify24.cu and sparse_gemm.cu
Deleted SRELU that depended on the above kernels
Deleted associated ops, references, and tests

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:
ghstack-source-id: d1537fa708e51cbc934788d886ba5d34a82e0ff0 Pull-Request: #3726

Summary: Deleted torchao/csrc/cuda/activation24
Included sparsify24.cu and sparse_gemm.cu
Deleted SRELU that depended on the above kernels
Deleted associated ops, references, and tests

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:
ghstack-source-id: d1537fa
Pull-Request: #3726
@pytorch-bot

pytorch-bot Bot commented Jan 28, 2026

Copy link
Copy Markdown

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/3744

Note: Links to docs will display an error until the docs builds have been completed.

⏳ No Failures, 1 Pending

As of commit 2695b13 with merge base 49c7a51 (image):
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

@meta-cla meta-cla Bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Jan 28, 2026
@howardzhang-cv howardzhang-cv added module: bc-breaking Use this tag if this PR breaks backward compatibility module: deprecation Use this tag if this PR deprecates a feature labels Jan 28, 2026
@howardzhang-cv howardzhang-cv merged commit a003c32 into main Jan 28, 2026
38 of 40 checks passed
jerryzh168 added a commit that referenced this pull request Apr 2, 2026
This benchmark was built around SRELUFloat8SemiSparseDynamicActivationFloat8WeightConfig
and SquaredReLU, both of which were deleted in #3744. All model definitions
in the file embedded SquaredReLU(), so no benchmark in the file was salvageable.

[ghstack-poisoned]
jerryzh168 added a commit that referenced this pull request Apr 2, 2026
This benchmark was built around SRELUFloat8SemiSparseDynamicActivationFloat8WeightConfig
and SquaredReLU, both of which were deleted in #3744. All model definitions
in the file embedded SquaredReLU(), so no benchmark in the file was salvageable.

ghstack-source-id: 452b516
Pull Request resolved: #4217
jerryzh168 added a commit that referenced this pull request Apr 2, 2026
This benchmark was built around SRELUFloat8SemiSparseDynamicActivationFloat8WeightConfig
and SquaredReLU, both of which were deleted in #3744. All model definitions
in the file embedded SquaredReLU(), so no benchmark in the file was salvageable.

[ghstack-poisoned]
Priyjain-amd pushed a commit to Priyjain-amd/ao that referenced this pull request May 26, 2026
This benchmark was built around SRELUFloat8SemiSparseDynamicActivationFloat8WeightConfig
and SquaredReLU, both of which were deleted in pytorch#3744. All model definitions
in the file embedded SquaredReLU(), so no benchmark in the file was salvageable.

[ghstack-poisoned]
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. module: bc-breaking Use this tag if this PR breaks backward compatibility module: deprecation Use this tag if this PR deprecates a feature

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants