pytorch

94b5c807 - Detach fake tensors into val, so they aren't affected by metadata mutation (#89140)

Detach fake tensors into val, so they aren't affected by metadata mutation (#89140) Signed-off-by: Edward Z. Yang <ezyang@fb.com> Pull Request resolved: https://github.com/pytorch/pytorch/pull/89140 Approved by: https://github.com/bdhirsh