Open kaixih opened 3 weeks ago
LGTM - please rebase and clear the CI test checks then I can merge it.
@IvyZX Thanks for the comments. Just resolved the conflict. PTAL.
Attention: Patch coverage is 0%
with 6 lines
in your changes missing coverage. Please review.
Project coverage is 0.00%. Comparing base (
31adb00
) to head (798cfe7
). Report is 59 commits behind head on main.
Files | Patch % | Lines |
---|---|---|
flax/linen/fp8_ops.py | 0.00% | 6 Missing :warning: |
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
This PR renames the original
fm32
tofp32_max_grad
to express the idea of the dtype is used for storing fp32 values and using max for the gradient accumulation.cc. @nouiz