If you’re not making factor bets then all of the active risk should be attributable to holding individuals securities in different weights or different from the benchmark - not risk from a sector focus, for example.
On # of stocks, as # increases then more likely holding more securities in the benchmark so lower active share and lower tracking error.