wenchenvincent

Results 8 comments of wenchenvincent

[ROCm] Added MIOpen header files to installation package for ROCm.

> @wenchenvincent please sign CLA so we can proceed. Thanks. Thanks Jeff! I have just signed the CLA.

value conversion bug for relatively large numbers

Thank you for the answer! Is there any option that we can choose to represent the numbers as double precision?

[Feature Request] Any roadmap for supporting FP8 attention calculation?

I think this commit added fp8 fused attention: https://github.com/NVIDIA/TransformerEngine/commit/989a53a06478a4223ffb2fc2fc92b5febcf9d8c1#diff-236e240f7f5506de96cfd5f61c77c7142905dabada33f6f0c68094724dbfb9b4

Added support of NANOO fp8.

@levskaya I noticed that you have reviewed several PRs regarding fp8. Could you take a look at this one?

Added support of NANOO fp8.

@levskaya Could you kindly serve as the reviewer for this PR?

Added support of NANOO fp8.

@levskaya Thanks for the review. I have updated the PR to address the concerns. Could you take a look at the updates?

Added support of NANOO fp8.

> Thanks for the fixes! We may need to do some tiny rebasing of simple things as the codebase just migrated to a python minver of 3.10. Thanks! Do you...

Added support of NANOO fp8.

> Yes to tip as of today should have the 3.10 minver updates. Also, I'm seeing this failure in the tests: > > ``` > FAILED tests/linen/linen_test.py::Fp8Test::test_fp8_meta_dtype0 - TypeError: missing...