Post by Bogdan Buduroiu
@jeridansky * sigh * can't wait to see the AP write "The sampled token sequence is consistent with a low-loss region of the policy manifold in which the KL-regularised objective, as shaped by RLHF, a…
@jeridansky * sigh * can't wait to see the AP write "The sampled token sequence is consistent with a low-loss region of the policy manifold in which the KL-regularised objective, as shaped by RLHF, a…