Of course you can't see a force but you can see that the 2 small spheres connected to the torsion fiber are moves a certain angle. There is a force needed to give an angle to the torsion fiber. This force is the force between the small and the big sphere. Because this is the only thing that has been changed during the experiment.
What changed during the experiment? I don't get it. And why are you assuming there is a force?
You, "TheEngineer" don't make me laugh.
As you can see on the images on
http://www.leydenscience.org/physics/gravitation/cavend.htm1) You have a torsion fiber with a mirror and 2 small lead spheres
2) You can place a laser and a screen so that the laser ray is orthogonal to the mirror and is reflecting to a point on the screen the screen.
3) You place the 2 big lead spheres, you'll now see that the laser ray is gong to another point on the screen.
Conclusion: The fiber is twisted, for twisting a fiber you'll need a force
=> in 3) you have an extra force comparing to 2). (a)
What has been changed during the experiment (2-3) (this are the possible reasons of the force): only one thing changed: 2 big lead spheres. (b)
(a) + (b)
=> The only thing that can cause the extra force is the addition of the 2 big lead spheres.
=> There must be a force between the big lead spheres and the small lead spheres.
With the law of gravitation you can predict the force. If you insert this force in the torsion equation you can predict the angle of twisting. Experiments give the same value as the equations (only some occasional errors, like you have in every experiment).
QED