Usually it's like this:
(input stage) -- (tone control passive circuit stage) --)| -- (power stage, where amplifier gains you dB/volts) -- |( -- (output stage)
Input, power, tone control, output.
Where |( and )| are polarized capacitors to stop the power stage from affecting the input and output.
You can imagine that the tone stack prior to a gain stage is going to be huge amplification of the Tonally modified signal, exaggerating the changes made to said signal. The tone stack after the gain stage means you get a far more delicate tonal change as you turn the potentiometers.
You could even do both if you wanted. A pre-gain stack and a post-gain stack, for massive and minute changes to signal.
Decoupling with the polarized caps is important, because it stops high voltage DC from using the tone stack circuits, provided you don't have ANY power rail + connection to those tone stacks. The ground on these circuits will deal with the current of the AC signal only.
If you make a mistake and have the tone stacks connected to the DC rail by a physical wire instead of an inductive gap like a polarized cap(I say they are inductive because One side charges and promotes charging of opposite voltage value on the other side, without any physical contact, which is the same way true inductors of coiled wire pass voltage to each other) , you'll be weakening the DC current and voltage through any gain stage (transistors), while also disrupting the intended design of the tone stack. It can do anything from not working to sounding terrible.
Building such circuits are much easier than true Active Tone Control, since ATCs reside within the power stage itself, which means you have to design around the possibility of the tone modification ruining biasing and/or causing unwanted clipping.
Tone circuits and how they work in a nutshell:
RC circuits, due to the reactance of Capacitors, which is their resistance or slowness to change when hit with voltage/current, cause phase shifts. The amount of phase shift for each frequency depends on some calculation of R and C.
Phase shifts cause a frequency in a signal to fight itself, which is how you get more/less treble or bass, etc. The amount of how much it fights itself (or not) determines how much it is attenuated.
At muzique.com, you can see pretty good examples of really simple circuits being "tone modifiers" via Treble booster and Bass booster. The size of the caps in conjunction with the resistors and TransformingResistors (transistors), are RC circuits themselves, which is why they have an affect on tone. I just recently built myself a Treble booster out of a 2n2222 single transistor in a common collector configuration. The gain in the 800hz+ region is pretty massive. It behaves like the Screaming bird treble booster, but it has greater boost out to mid-range, and plateaus sooner so it seems like it'd sound pretty decent and fuller.
I've only been doing this 4 to 5 months, so if there's any mistakes, the others will probably correct it.