Why are these models able to reduce parameters but keep quality? I know the original intuition was scale data + params = quality but it looks like we have hit an s curve on improvements from pure scaling? Is this just because we are in a memory / data crunch? Are we learning how LLMs learn and effectively training better? Do we have a way to derive the amount of intelligence an LLM will have based on size / training / etc that isn't just brute force ablations?

Presumably there is distillation or similar being used to transfer from a larger model to a smaller one.

They innovated a lot.