"Basically I want to see what is going on, but be able to jump in if a traceback occurs."
Here's a radical thought: Don't.
"Watching" is a crutch. You should be writing small sections of code that you know will work. Then put those together.
Watching sometimes results from "I'm not sure what Python's really doing", so there's an urge to "watch" execution and see what's happening. Other times watching results from writing a script that's too big and complex without proper decomposition. Sometimes watching results from having a detailed specification which was translated into Python without a deep understanding. I've seen people doing these; of course, there are lots more reasons.
The advice, however, is the same for all:
Break things into small pieces,
usually classes of functions. Make
them simple enough that you can
actually understand what Python is
doing.
Knit them together to create your
larger application from pieces you
actually understand.
Watching will limit your ability to actually write working software. It will -- in a very real way -- limit you to trivial programming exercises. It's not a good learning tool; and it's a perfectly awful way to create production code.
Bottom Line.
Don't pursue "watching". Decompose into smaller pieces so you don't need to watch.