Pinned
Our key assumption? LLMs don’t actually exist at the level of discrete predictions, but in the gradient field of possibilities that sits behind each token.
Anybody that tells you we understand what is going on inside an LLM during training is deluding you or themselves. They
Replying to @AnthropicAI
Our key assumption? We imagine that evaluation questions are randomly drawn from an underlying distribution of questions. This assumption unlocks a rich theoretical landscape, from which we derive five core recommendations.
