Anthropic can now track the inner workings of a large language model

Forgive my ignorance but is it possible to visualize neural net activation during inference?