Update op replacement tutorial (#50377)

Summary: Pull Request resolved: #50377

Test Plan: Imported from OSS

Reviewed By: jamesr66a

Differential Revision: D25870409

Pulled By: ansley

fbshipit-source-id: b873b89c2e62b57cd5d816f81361c8ff31be2948

torch/fx/examples/replace_op.py

@@ -2,7 +2,6 @@
 from torch.fx import symbolic_trace
 import operator
 
-
 """
 How to replace one op with another
 1. Iterate through all Nodes in your GraphModule's Graph.
@@ -39,12 +38,10 @@ def forward(self, x, y):
 
 # As demonstrated in the above example, there are several different ways
 # to denote addition. The possible cases are:
-#     1. `x + y` - A `call_function` Node with target
-#        `<built-in function add>`. This is `operator.add`, so we can
-#         match on equality with that function directly.
+#     1. `x + y` - A `call_function` Node with target `operator.add`.
+#         We can match for equality on that `operator.add` directly.
 #     2. `torch.add(x, y)` - A `call_function` Node with target
-#        `<built-in method add of type object at MEMORY-LOCATION-OF-TORCH>`.
-#         This is `torch.add`, which we can similarly match directly.
+#         `torch.add`. Similarly, we can match this function directly.
 #     3. `x.add(y)` - The Tensor method call, whose target we can match
 #         as a string.