Inside a crystal, electrons act weirdly like an electron that travels with a "perturbation" around it, so the mass for calculations is not the mass of an electron traveling in vacuum. It's call "effective mass" m* https://en.wikipedia.org/wiki/Effective_mass_(solid-state_ph...

The weird part is that it can be negative, so we already know quasi-particles with negative mass.

This quasi particle with negative mass and negative charge is a fermion so it's mathematically equivalent to a quasi particle with positive mass and positive charge. We call them "holes" and people that work with semiconductors think about "holes" instead of weird particles with negative mass.

Anyway, inside semiconductors they slow down and follow the conservation of energy law, they don't get faster and faster like in the simulation. The weird behavior of the simulation is not realistic.