Fitness Function Rating: Accuracy based on user defined tolerance level.  This is the value that you should use to determine if network has trained long enough.

Current error: Value that the network computes as to provide a means of self monitoring.  As this approaches zero network should become fully trained.

Best error: Value captured by training manager.  This value may be reset at any time by the training manager if it is determined that network has hit a local minimum

Momentum and Learning rate values: Values are indicators of how much impact a training iteration has on network. These values are automatically modified by training manager unless locked.