ar0cket1
@ar0cket1
not enough information | RL science
244 Following 1.6K Followers
really cool, didn't know @oneill_c was Australian until i heard the accent
I hadn't really considered the prompt distribution and the usefulness of realistic data in that regard and how the distillation goes through these channels, its a good explanation of why Chinese models are so good compared to like sonnet or opus though.
Show more
dont create architectures that become harder to RL (cough cough continuous models that use previous latent states as the first state of the next generated token), create archiectures that are built to be RLed.
Show more
