[Fix typo] Fix typo error in grad_node_info.h (#51607)

* [Fix typo] Fix typo error in grad_node_info.h

* fix varbase_patch_methods.py typo error

* fix more typo errors

paddle/fluid/eager/general_grad.h

@@ -31,7 +31,7 @@
 namespace egr {
 
 /*
- * GeneralGrad is Helpper class to implement custom grad operation between
+ * GeneralGrad is Helper class to implement custom grad operation between
  * outputs and inputs.
  *
  * **/

paddle/fluid/eager/grad_node_info.h

@@ -25,7 +25,7 @@ namespace egr {
 /**
  * GradNodeBase is base class of all grad node, which is what should be used by
  * eager execution, we define most of backward autograd members here, and for
- * each Operator, they should hold their onw forward Inputs as TensorWrapper.
+ * each Operator, they should hold their own forward Inputs as TensorWrapper.
  *
  * The GradNodeBase will be held in autograd_meta, and it is also a member of
  * Edge, which indicates the edge of backward graph.
@@ -40,7 +40,7 @@ namespace egr {
  *
  * NOTE: GradNodeBase holds its own inputs and Outputs
  *
- * Edge is defined to descripe depend of backward, an Edge is what linked
+ * Edge is defined to describe depend of backward, an Edge is what linked
  * between two node, it should contain a Node and rank of this Node (this is
  * used to indicate which input of grad this edge belong).
  **/

python/paddle/fluid/dygraph/varbase_patch_methods.py

@@ -232,7 +232,7 @@ def monkey_patch_varbase():
         """
         Run backward of current Graph which starts from current Tensor.
 
-        The new gradient will accumulat on previous gradient.
+        The new gradient will accumulate on previous gradient.
 
         You can clear gradient by ``Tensor.clear_grad()`` .
 
@@ -240,11 +240,11 @@ def monkey_patch_varbase():
             grad_tensor(Tensor, optional): initial gradient values of the current Tensor. If `grad_tensor` is None,
             the initial gradient values of the current Tensor would be Tensor filled with 1.0;
             if `grad_tensor` is not None, it must have the same length as the current Tensor.
-            Teh default value is None.
+            The default value is None.
 
             retain_graph(bool, optional): If False, the graph used to compute grads will be freed. If you would
                 like to add more ops to the built graph after calling this method( :code:`backward` ), set the parameter
-                :code:`retain_graph` to True, then the grads will be retained. Thus, seting it to False is much more memory-efficient.
+                :code:`retain_graph` to True, then the grads will be retained. Thus, setting it to False is much more memory-efficient.
                 Defaults to False.
         Returns:
             NoneType: None