very small typo in fsdp2 comment (pytorch#163155)

Pull Request resolved: pytorch#163155
Approved by: https://github.com/awgu, https://github.com/Skylion007

Author: zhc7

torch/distributed/fsdp/_fully_shard/_fsdp_state.py

@@ -230,7 +230,7 @@ def _pre_forward(
         self, module: nn.Module, args: tuple[Any, ...], kwargs: dict[str, Any]
     ) -> tuple[tuple[Any, ...], dict[str, Any]]:
         # When composing with module-hook-based activation checkpointing, the
-        # the pre-backward hook is responsible for the unshard
+        # pre-backward hook is responsible for the unshard
         if self._training_state == TrainingState.PRE_BACKWARD:
             return args, kwargs
         self._training_state = TrainingState.FORWARD