Circuit Tracing: Revealing Computational Graphs in Language Models

• https://transformer-circuits.pub/2025/attribution-graphs/methods.html
• Emmanuel Ameisen, Jack Lindsey, … Chris Olah, Joshua Batson

Interpretability, Model interpretability

See also Lindsey2025on. Early works published in Distill: Thread: Circuits, which includes the following articles:

• Olah2020zoom

Replacement model and Attribution graph are core concepts.