jacob_drori comments on It Is Reasonable To Research How To Use Model Internals In Training

[jacob_drori]

My guess is doing it with RL is worse because I expect it to generalize to more probes

Why?