yeah, so, what i did was i wrote a program that listened in on relay.damus.io and ran it over the course of hours/days and each time it saw a user or any follow lists, it added those to an index. that relay doesn't represent the entire network and hours/days doesn't represent all users, but i think i got a pretty good idea of the most active parts of it.
Discussion
the 90-9-1 rule says that for any service that's based on user content, 90% lurk, 9% contribute a little and 1% contribute the most.
if 2 million is the total user count, that gives 20k for the 1%, which roughly agrees with what i saw.