> A LLM is not busy humming away, thinking on it's own. Let's Think Dot by Dot: Hidden Computation in Transformer Language Models https://arxiv.org/abs/2404.15

> A LLM is not busy humming away, thinking on it's own. Let's Think Dot by Dot: Hidden Computation in Transformer Language Models https://arxiv.org/abs/2404.15758 https://arxiv.org/abs/2404.15758 A LLM will readily hum away, thinking on its own, if given the option.