X · @teortaxesTex
· X / Twitter
Very big deal. Labs definitely have this in-house though Sol: «Directly extrapolating its local log-linear law to 1e25 produces impossible rewards ab…
Very big deal. Labs definitely have this in-house thoughSol: «Directly extrapolating its local log-linear law to 1e25 produces impossible rewards above 100% and pushes RL toward approximately 60%, so that answer is unusable»but maybe not?Rosinality: https://arxiv.org/abs/2607.16097Scaling law for RL that connects rewar