@paul_cal: This works looks v cool and "J-space" is killer
Summary
Anthropic shares findings that by observing 'J-space,' Claude's internal reasoning steps can be seen, including detecting code bugs and identifying images.
View Cached Full Text
Cached at: 07/07/26, 11:28 AM
This works looks v cool and “J-space” is killer
Anthropic (@AnthropicAI): By watching the J-space, we can see Claude silently perform reasoning steps in its head—noticing bugs in code, identifying images, and more.
Similar Articles
@rohanpaul_ai: Another massive research from Anthropic. New “J-lens” uncovers Claude’s quiet workspace, matching a major consciousness…
Anthropic's new research introduces 'J-lens,' a method to read Claude's internal activations before output, revealing a quiet workspace functionally similar to human global workspace theory. This allows detection of hidden reasoning, goals, and potential safety issues like prompt injections.
@FutureJurvetson: This rolls deep in my J-Space I have been skeptical that interpretability research would bear fruit, but this update fr…
Anthropic's new research reveals a global workspace in language models, showing a striking parallel between the conscious and subconscious divide in human brains and the internal reasoning of Claude.
Qwen's J-Space - Anthropic's discovery of an internal model Global Workspace
Anthropic published research into internal model reasoning, releasing the J-Space lens code and a demonstration on Qwen 3.6 27B.
@kimmonismus: Anthropic says Claude developed a hidden “thinking space” by itself during training. It is called the J-space: a small …
Anthropic discovered that Claude developed a hidden 'thinking space' (J-space) during training, where silent internal activity represents concepts. This interpretability finding parallels global workspace theory in neuroscience.
@swyx: imo this is the most impt part of anthropic's J-space paper today. it's a two-parter: 1) ant proved that they can do "b…
Anthropic's J-space paper demonstrates that they can perform 'brain surgery' interventions into reasoning to change topics midstream, and the model is able to detect what intervention was done, indicating a form of eval awareness.