Hardcoded tuning for FBGEMM/fp8 for 70B/405B Prefill with T=1..8K (#3… #5190
facebook-github-botfbgemm_gpu_ci_rocm.yml
on: push
Matrix: build_artifact
Matrix: test_and_publish_artifact
Annotations
2 warnings
|
test_and_publish_artifact (x86, rocm, 3.12, 6.1, gcc)
The following actions use a deprecated Node.js version and will be forced to run on node20: actions/checkout@v3. For more info: https://github.blog/changelog/2024-03-07-github-actions-all-actions-will-run-on-node20-instead-of-node16-by-default/
|
|
test_and_publish_artifact (x86, rocm, 3.12, 6.1, clang)
The following actions use a deprecated Node.js version and will be forced to run on node20: actions/checkout@v3. For more info: https://github.blog/changelog/2024-03-07-github-actions-all-actions-will-run-on-node20-instead-of-node16-by-default/
|
Artifacts
Produced during runtime
| Name | Size | |
|---|---|---|
|
fbgemm_gpu_nightly_rocm_x86_clang_py3.10_rocm6.1.whl
|
53.2 MB |
|
|
fbgemm_gpu_nightly_rocm_x86_clang_py3.11_rocm6.1.whl
|
53.2 MB |
|
|
fbgemm_gpu_nightly_rocm_x86_clang_py3.12_rocm6.1.whl
|
53.2 MB |
|
|
fbgemm_gpu_nightly_rocm_x86_clang_py3.9_rocm6.1.whl
|
53.2 MB |
|
|
fbgemm_gpu_nightly_rocm_x86_gcc_py3.10_rocm6.1.whl
|
53.3 MB |
|
|
fbgemm_gpu_nightly_rocm_x86_gcc_py3.11_rocm6.1.whl
|
53.3 MB |
|
|
fbgemm_gpu_nightly_rocm_x86_gcc_py3.12_rocm6.1.whl
|
53.3 MB |
|
|
fbgemm_gpu_nightly_rocm_x86_gcc_py3.9_rocm6.1.whl
|
53.3 MB |
|