BREAKING
Anthropic finds 'J-space' in Claude
A quiet internal workspace
0%
of activations
0
Sonnet version
How the Jacobian Lens works
1Scan vocabulary
2Find activation patterns
3Apply layer by layer
4Watch J-space shift
What it means, and what it doesn't
Safety potentialupside
Detect prompt injection
Visualize reasoning steps
Big caveatcaution
Access, not phenomenal
Still research stage
Open-sourced for verification
AI NEWS BLITZ
Anthropic says it found a hidden thinking space inside its Claude models.