Sagittarius A*, the closest black hole, is 4.1 million solar masses, so its Schwarzchild radius is 12.1 million km, so its density is 0.25 kg/m^3, which is about 1/5 as dense as air at sea level--much less than water.
A black hole made from Jupiter would have a Schwarzchild radius of 2.8 meters, so its density would be... denser than a neutron star.
If you look into holographic theory (which I prefer personally) then the entire mass of a black hole is represented as information on the surface of the black hole, explaining the relation between mass and surface not mass and volume. This solves some problems with the information theorem (no information can be lost) by presenting all information contained within a black hole on it's surface where it can be read, atleast theoretically. (Here information and mass would be considered equivalent)
To some extend, it still holds true if you just consider the mass of a black hole to be it's singularity and it's volume the volume that fills the event horizon.
edit: Also consider; if you placed an equivalent amount of mass in space that occupies less volume than the event horizon would, it would automatically be a black hole (and note; the schwarzschild radius is not necessarily the event horizon, the event horizon can be tighter if the black hole rotates)
If it grew more than linearly with mass, we wold see the same reducing density, just faster.
It's an interesting factoid that blackhole mass isn't related to their volume but their surface area, so by adding mass they grow a lot faster than you'd expect and density goes down fast. Larger black holes are barely denser than water.