Anthropic has developed a tool to reveal the internal reasoning steps of its Claude model, which appears to resemble human brain function. The tool, referred to as J-lens or Jacobian Lens, is said to expose a hidden workspace or "J-space" within the model, allowing insight into its silent reasoning process. The discovery has been likened to a leading theory of consciousness, with some outlets describing it as revealing the model's "hidden mind" or "inner monologue".