Tag
2 posts tagged #interpretability
Browse 2 posts tagged Interpretability, including practical setup notes, reviews, comparisons, and workflow patterns for engineers working with AI tools.
A free browser tool showing which internal concepts a language model holds mid-computation, using Anthropic's Jacobian lens to let you observe reasoning layer by layer.
A free web tool that visualizes internal activation patterns of language models layer by layer as they process a prompt, letting you observe and manipulate concepts mid-computation.