Neuronpedia – Scoping out LLM-brains.
That is quite literally what you can do with a demo from Neuronpedia, a platform for AI interpretability. Whats that again? Mechanistic Interpretability is the field of research focused on understanding the inner workings of AI.