Apply suggestions from code review

Co-authored-by: Nick Robinson <npr251@gmail.com>

Author: oxinabox

README.md

@@ -12,5 +12,5 @@
 [![](https://img.shields.io/badge/docs-master-blue.svg)](https://juliadiff.org/ChainRulesOverloadGeneration.jl/dev)
 [![](https://img.shields.io/badge/docs-stable-blue.svg)](https://juliadiff.org/ChainRulesOverloadGeneration.jl/stable)
 
-The ChainRulesOverloadGeneration package provides a suite of methods for using [ChainRulesCore.jl](https://github.com/JuliaDiff/ChainRulesCore.jl) rules in operator overloaded based AD systems.
+The ChainRulesOverloadGeneration package provides a suite of methods for using [ChainRulesCore.jl](https://github.com/JuliaDiff/ChainRulesCore.jl) rules in operator overloading AD systems.
 It tracks what rules are defined at any point in time, and lets you trigger functions to which can use `@eval` in order to define the matching operator overloads.

docs/src/examples/forward_mode.md

@@ -1,6 +1,6 @@
 # ForwardDiffZero
 This is a fairly standard operator overloading-based forward mode AD system.
-It defines a `Dual` part which holds both the primal value, paired with the partial deriviative.
+It defines a `Dual` part which holds both the primal value, paired with the partial derivative.
 It doesn't handle chunked-mode, or perturbation confusion.
 The overload generation hook in this example is: `define_dual_overload`.
 
@@ -12,4 +12,3 @@ $(read(joinpath(@__DIR__,"../../../test/demos/forwarddiffzero.jl"), String))
 ```
 """)
 ````
-

docs/src/examples/reverse_mode.md

@@ -1,8 +1,8 @@
-
 # ReverseDiffZero
+
 This is a fairly standard operator overloading based reverse mode AD system.
-It defines a `Tracked` type which carries the primal value as well as a reference to the tape which is it using, a partially accumulated partial derivative and a `propagate` function that propagates it's partial back to its input.
-A perhaps unusual thing about it is how little it carries around it's creating operator's inputs.
+It defines a `Tracked` type which carries the primal value as well as a reference to the tape which is it using, a partially accumulated partial derivative and a `propagate` function that propagates its partial back to its input.
+A perhaps unusual thing about it is how little it carries around its creating operator's inputs.
 That information is all entirely wrapped up in the `propagate` function.
 The overload generation hook in this example is: `define_tracked_overload`.