[DOC][Train] Fix typo for Instantiating in ray train doc (#55826)

Signed-off-by: Potato <tanxinyu@apache.org>

Author: OneSizeFitsQuorum

doc/source/train/user-guides/data-loading-preprocessing.rst

@@ -322,7 +322,7 @@ For more details, see the following sections for each framework:
 .. tip::
 
     When using Torch or Hugging Face Datasets directly without Ray Data, make sure to instantiate your Dataset *inside* the ``train_loop_per_worker``.
-    Instatiating the Dataset outside of the ``train_loop_per_worker`` and passing it in via global scope
+    Instantiating the Dataset outside of the ``train_loop_per_worker`` and passing it in via global scope
     can cause errors due to the Dataset not being serializable.
 
 .. note::