Mechanistic Circuit Analysis at Frontier Scale: cells as a unit of interpretability

Abstract

We adapt mechanistic interpretability tooling to operate at the parameter scale of frontier-class models without quadratic costs in attention-head enumeration. Our core observation: features cluster into ~700 reusable circuits whose composition explains 86% of behaviourally relevant activations on the benchmarks we tested. The methodology is now used as a precondition for the closed-loop interpretability cells paired with every RSI run.

Index metadata

Cell

hilbert-13

Compute

92 H100-days

Status

Open release

Tool

ab-circuits open-source release

DOI

10.48550/arXiv.2505.89369

arXiv

arXiv:2505.89369

What this paper is part of

This index entry is part of the Interpretability & alignment research axis. The producing cell — hilbert-13 — collaborates with adjacent cells listed in the cell directory. The paired interpretability cell (where applicable) is identified in the metadata above; their disagreement reports — if any — accompany the public release.

How to read this

If you want to use the result: the code (where available) is at https://github.com/alphabell-labs/ab-mechanis; the dataset is at TBD when one is released. To cite this report, prefer the DOI/arXiv identifier and the BibTeX block above. To discuss this with the producing cell, contact the lab with the index entry slug mechanistic-circuits-frontier.

Limitations

Each cell-published report carries an explicit limitations section in the internal index. We do not paraphrase it here. Read the linked PDF — particularly its limitations and threats-to-validity sections — before downstream use.