fwd_rng_state show up in the aot_export_joint grpah input
Open
module: aotdispatch
module: pt2-dispatcher
oncall: pt2
triaged
- Dominant language
- Python
- Stars
- 103k
- Forks
- 29.5k
- PR merge metrics
- PR metrics pending
Description
See https://github.com/pytorch/torchtitan/pull/1794
P1975157784: rank0_autograd_function_0fea2786.py
Setting `torch._functorch.config.graphsafe_rng_functionalization = False` doesn't work.
How to avoid `fwd_rng_state` from showing up?
cc @chauhang @penguinwu @zou3519 @bdhirsh
Contributor guide
Assessment
This issue has not been assessed yet.