single mention · post-training
Moonlight scaling
A named scaling technique that was used in some RL runs and deactivated in another; its mechanism is not specified by the evidence.
The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See reinforcement learning algorithm for the methods filed beside it.
Also called Moonlight scaling in RL.
Filed under post-training › reinforcement learning algorithm