SOURCE-LINKED INTELLIGENCE

How Do Linear Probes Emerge? A Circuit-Tracing Framework with Concept-Targeted Attribution

arXiv · AI, language, vision and robotics · article · Aug 27, 2026 · UTC

Transcoder attribution graphs are usually trained to explain why a model assigns high probability to a particular next token. We introduce Concept-Targeted Attribution (CTA), which instead trains attribution graphs with respect to a linear probe direction. CTA therefore yields probe-specific circuits that explain why an internal concept representation arises in a prompt, independently of whether it is expressed in the generated token. Using Cross-Layer Transcoders, we show that these probe-targeted graphs contain predictive structure: graph-level features predict probe accuracy across four wid

Read original source ↗ Open in workspace

recordType

paper

region

Global