Jessica Rumbelow

Jessica Rumbelow

We introduce Exemplar Partitioning (EP), an unsupervised method for constructing interpretable feature dictionaries from large language model activations with $\sim 10^{3}\times$ fewer tokens than comparable sparse autoencoders (SAEs). An EP dictionary is a Voronoi partition of activation space, built by leader-clustering streamed activations within a distance threshold. Each region is anchored by an observed exemplar that serves as both its membership criterion and intervention direction; dictionary size is not prespecified, but determined by the activation geometry at that threshold. Because exemplars are observed rather than learned, dictionaries built from the same data stream are directly comparable across layers, models, and training checkpoints. We characterise EP as an interpretability object via targeted demonstrations of properties newly accessible through this construction, plus one head-to-head benchmark. In Gemma-2-2B, EP dictionary regions are interpretable and support causal interventions: refusal in instruction-tuned Gemma concentrates in a region whose exemplar ablation can collapse held-out refusal. Cross-checkpoint matching between base and instruction-tuned dictionaries separates the directions preserved through finetuning from those introduced by it. EP regions and Gemma Scope SAE features decompose activation space differently but agree on a shared core: $\sim 20\%$ of EP regions match an SAE feature at $F_{1} > 0.5$, and EP one-hot probes retain $\sim 97\%$ of raw-activation probe accuracy at $\ell_{0} = 1$. Nearest-exemplar distance provides a free out-of-distribution signal at inference. On AxBench latent concept detection at Gemma-2-2B-it L20, EP at $p_{1}$ reaches mean AUROC $0.881$, $+0.126$ over the canonical GemmaScope SAE leaderboard entry and within $0.030$ of SAE-A's $0.911$, at $\sim 10^{3}\times$ less build compute.

Lee Sharkey, Bilal Chughtai, Joshua Batson et al. · deepmind

Mechanistic interpretability aims to understand the computational mechanisms underlying neural networks' capabilities in order to accomplish concrete scientific and engineering goals. Progress in this field thus promises to provide greater assurance over AI system behavior and shed light on exciting scientific questions about the nature of intelligence. Despite recent progress toward these goals, there are many open problems in the field that require solutions before many scientific and practical benefits can be realized: Our methods require both conceptual and practical improvements to reveal deeper insights; we must figure out how best to apply our methods in pursuit of specific goals; and the field must grapple with socio-technical challenges that influence and are influenced by our work. This forward-facing review discusses the current frontier of mechanistic interpretability and the open problems that the field may benefit from prioritizing.

Arush Tagade, Jessica Rumbelow

We introduce Prototype Generation, a stricter and more robust form of feature visualisation for model-agnostic, data-independent interpretability of image classification models. We demonstrate its ability to generate inputs that result in natural activation paths, countering previous claims that feature visualisation algorithms are untrustworthy due to the unnatural internal activations. We substantiate these claims by quantitatively measuring similarity between the internal activations of our generated prototypes and natural images. We also demonstrate how the interpretation of generated prototypes yields important insights, highlighting spurious correlations and biases learned by models which quantitative methods over test-sets cannot identify.

Stephen Casper, Jieun Yun, Joonhyuk Baek et al.

Interpretability techniques are valuable for helping humans understand and oversee AI systems. The SaTML 2024 CNN Interpretability Competition solicited novel methods for studying convolutional neural networks (CNNs) at the ImageNet scale. The objective of the competition was to help human crowd-workers identify trojans in CNNs. This report showcases the methods and results of four featured competition entries. It remains challenging to help humans reliably diagnose trojans via interpretability tools. However, the competition's entries have contributed new techniques and set a new record on the benchmark from Casper et al., 2023.

Jack Foxabbott, Arush Tagade, Andrew Cusick et al.

The Discovery Engine is a general purpose automated system for scientific discovery, which combines machine learning with state-of-the-art ML interpretability to enable rapid and robust scientific insight across diverse datasets. In this paper, we benchmark the Discovery Engine against five recent peer-reviewed scientific publications applying machine learning across medicine, materials science, social science, and environmental science. In each case, the Discovery Engine matches or exceeds prior predictive performance while also generating deeper, more actionable insights through rich interpretability artefacts. These results demonstrate its potential as a new standard for automated, interpretable scientific modelling that enables complex knowledge discovery from data.