Resolution of Kleinian Singularities using Hilbert schemes of points

Apologies in advance for the naive and rather speculative question.

In this blog post by John Baez, he paints a (perhaps not original) picture of how one might expect that the minimal resolution of a Kleinian singularity $$\mathbb{C}^2/G$$ is given by Nakajima's equivariant Hilbert scheme of points. At the end though, he admits that its not clear how one could interpret the exceptional divisor, indeed the end of the post reads:

As I hope you see, these are certain ‘limits’ of 600-cells that have ‘shrunk to the origin’… or in other words, highly symmetrical ways for 120 points in C2 to collide at the origin, with some highly symmetrical conditions on their velocities, accelerations, etc.

That’s what I need to understand.

I am looking for some resource or a pointer to how one could formalize this further. While both the Hilbert scheme and the blow up are crepant resolutions, he seems to be trying to understand the exceptional divisor from the viewpoint that the resolution is a "parameter space" (informally) for various positions of the polyhedra that the finite subgroup acts on. My interest is exactly this viewpoint; rather then blowing up, as computing the intersection graph of the exceptional divisor seems like a "coincidence" to me, I think this viewpoint could help me understand the McKay correspondence better.