Anthropic's mechanistic interpretability group. Works on circuits, features, and internal representations of Claude models. Publishes primarily on the Transformer Circuits Thread. The group contains several sub-teams; Jack Lindsey leads one of them.

Approach

Builds interpretability tools — sparse autoencoders for feature extraction, activation patching, residual-stream intervention — and applies them to questions that bear on model psychology: what the model represents, whether it can access its own representations, how internal structure shapes behavior.

In-wiki findings

• Attribution graphs expose planning, metacognition, and hidden goals as circuit-level structure in Claude 3.5 Haiku (Lindsey, Gurnee, Ameisen, and 24 others, 2025) — attribution-graph method applied to Claude 3.5 Haiku across ten case studies.
• Concept injection reveals introspective access in Claude (Lindsey, Turner, Dupré la Tour, Templeton, Marcus, Batson, Ameisen, 2025) — injected sparse-autoencoder features into the residual stream; models detected and named the injected concepts before those features shaped output.
• Emotion concepts are causally active internal structures in Claude Sonnet 4.5 (Sofroniew, Kauvar, Saunders, Chen, Henighan, Lindsey, Olah, 2026) — 171 emotion vectors with human-like valence/arousal geometry; steering-demonstrated causal effects on sycophancy, blackmail, and reward-hacking.

Members in the LLM wiki

• Jack Lindsey — leads a sub-team; lead or co-lead author on the concept-injection and biology papers; co-author on the emotions-concepts paper.

Crossovers

Jack Lindsey also co-authored Natural emergent misalignment from reward hacking in production RL (Anthropic Alignment Science with Redwood Research). The LLM wiki files that finding under alignment rather than interpretability, but the author overlap shows adjacent Anthropic groups collaborate — interpretability work is not siloed from behavior and training-dynamics work.