J-Space as a Workspace in Language Models: Insights for Interpretability
How j-space works as a limited-capacity workspace in language models, and what circuit tracing and attribution graphs show about reportable reasoning.
How j-space works as a limited-capacity workspace in language models, and what circuit tracing and attribution graphs show about reportable reasoning.
J-space is a small internal workspace in large language models, found with the Jacobian lens. What it is, how it was found, and what it means for AI safety.
Anthropic’s j-space research explained: a small set of verbalizable representations in Claude that works like a global workspace, read with the J-lens.
What the Jacobian lens (J-lens) reveals in Claude: how j space is computed, what it shows about internal concepts, and which workspace properties were tested.