The paper introduces Exemplar Partitioning (EP), an unsupervised technique that constructs interpretable feature dictionaries from large language model activations by clustering streamed activations into Voronoi regions defined by exemplars and their averages. EP allows comparison of dictionaries across layers, checkpoints, and architectures, and demonstrates utility in interpreting model behavior, tracking training dynamics, detecting hidden concepts, and enabling targeted interventions. Experiments on Gemma‑2‑2B and Llama‑3.1‑8B show EP can reveal how instruction tuning reorganizes harmful prompt activations, facilitate interventions that alter model responses, and achieve high concept‑detection performance while requiring far fewer construction tokens than comparable methods.
By Jessica Rumbelow
arXiv:2606. 12629v1 Announce Type: cross Abstract: We show that the standard basis of transformer hidden states already provides a training-free, architecture-general feature basis.
By Varun Reddy Nalagatla
MECHVAR is a lightweight, auditable rule for selecting experiments from a finite library to discriminate between candidate mechanisms. It chooses probes by maximizing the posterior‑weighted variance of predicted responses, a score that aligns with the Box–Hill pairwise‑KL criterion and links to expected information gain when separations are small. Experiments on a 25‑block audit and a Digits loop show MECHVAR outperforming confirmation‑first strategies and matching or exceeding EIG in identification accuracy while being far faster to compute.
By Yifan Guo
arXiv:2607. 19364v2 Announce Type: replace Abstract: Activation steering adds a residual-stream direction at inference time, providing lightweight behavioral control without fine-tuning.
By Oshayer Siddique, J. M Areeb Uzair Alam, Md Jobayer Rahman Rafy, Syed Rifat Raiyan, Hasan Mahmud, Md Kamrul Hasan
arXiv:2605. 27259v2 Announce Type: replace Abstract: We propose Kan Extension Transformers (KETs) as a categorical design language for a diverse group of Transformer implementations.
By Sridhar Mahadevan
arXiv:2606. 14990v1 Announce Type: cross Abstract: Sparse autoencoders (SAEs) are standard tools for mechanistic interpretability, but current SAE families are constrained by fixed encoder nonlinearities such as ReLU, JumpReLU, and TopK.
By Naiyu Yin, Yue Yu
The paper reports that in English all‑words word sense disambiguation (WSD), the scarcity of high‑quality labels—not the models—has become the limiting factor. The authors introduce lexEN, a human‑adjudicated correction layer over the Maru2022 ALL_NEW benchmark, and SenseBench, a living leaderboard for LLM WSD evaluation. They show that frontier large language models reach about 95 % accuracy on lexEN‑v1, that relabeling corpora with these models improves downstream systems, and that fine‑grained WordNet senses are often ill‑posed, with coarsening improving both annotator agreement and model performance.
"whyItMatters":"The study highlights that improving label quality and managing annotation costs are now the critical challenges for advancing WSD performance, as model accuracy is already near its theoretical ceiling."
By Vassili Philippov, Amro Salman, Dmitrii Andreev, Penny Hands, Emil Kaiumov, Pavel Katunin, Anton Nikolaev
arXiv:2609.18961v1 Announce Type: new
Abstract: Mechanistic interpretability identifies sparse subsets of heads and MLP blocks that carry specific behaviors. We ask whether such causal signals can gu...
By Son Ha Xuan, Phat T. Tran-Truong, Xuan-Bach Le
arXiv:2603. 04198v2 Announce Type: replace-cross Abstract: Sparse autoencoders (SAEs) are widely used to extract human-interpretable features from neural network activations, but their learned features can vary substantially across random seeds and training choices.
By Piotr Jedryszek, Oliver M. Crook
arXiv:2609. 16391v1 Announce Type: cross Abstract: Weight-only post-training quantization is the cheapest way to shrink a retrieval embedder, and the received advice for applying it -- protect the embedding table, allocate bits by module sensitivity, prefer a ranking-aware objective over weight reconstruction -- was carried into LLM quantization largely intact.
By Hyojung Han
AMLworld-Compact introduces an importance‑weighted downsampling protocol for evaluating large language models on AMLworld’s test splits, retaining all illicit transactions while sampling benign ones from supervised‑score strata. The HT‑Coreset method reduces evaluation targets by 271× on HI‑Small and 611× on LI‑Small, providing unbiased confusion‑count estimates with exact recall and ratio estimates for precision and F1. Experiments with boosted‑tree strata and Graph Feature Preprocessor features show lower F1 RMSE compared to uniform sampling, and audit tools reveal that lexical checks do not reliably indicate correctness of model outputs.
By Natapong Nitarach, Phume Ngampornsukswadi, Pittawat Taveekitworachai, Surapon Nonesung, Warit Sirichotedumrong, Kunat Pipatanakul
arXiv:2609. 12591v1 Announce Type: new Abstract: Foundation models are increasingly adapted through fine-tuning, model editing, and alignment procedures while retaining previously acquired capabilities.
By Hendrik Droste, Christian Medeiros Adriano, Kathrin Korte, Holger Giese