Hey, I usually test my code by commenting out old versions, and my files sometimes get really long because of all the commented lines and detailed function notes I add. This causes the token count to skyrocket, and the model can’t process the entire file, which affects its accuracy.
I know I can set a rule to ignore commented lines, but that doesn’t really help because the model still counts those lines as tokens. Could we add an option to completely ignore commented lines when attaching the file? This would make it easier to work with longer files and focus on the actual code