← Back to all news
arXiv AI By Runyu Wang, Bo Liu, Xiaxin Zhang, Yu Han, Jiawei Cao, Xiaoye Zhang, Zhe Zhang, Yifan Yang, Peng Ping

RACE: Scalable Statistical Estimation of Functional Consistency in LLM Neurons

Read the original on arXiv AI →

RACE (Residual Alignment for Consistency Estimation) is a forward‑pass statistical framework that assesses the domain‑wide functional consistency of Transformer neurons. It outperforms gradient‑based point estimates in terms of domain specificity and shows a clear association between selected neurons and target domains through token‑distribution experiments. Importantly, RACE’s computational overhead is two orders of magnitude lower than that of gradient‑based methods, enabling scalable domain‑wide analysis.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

Related stories

arXiv Machine Learning
Oct 22

SYNAPSE: Framework for Neuron Analysis and Perturbation in Sequence Encoding

SYNAPSE is a training‑free framework that probes Transformer models by extracting per‑layer [CLS] representations, ranking neurons with a lightweight linear probe, and applying forward‑hook interventions during inference. It enables controlled, repeatable experiments on internal representations without altering the original model, revealing that task‑relevant information is encoded in broad, overlapping neuron subsets and that small weight or logit manipulations can redirect predictions. The framework highlights functional stability, class‑specific specialization, and complementary vulnerability profiles across domains.

By Jes\'us S\'anchez Ochoa, Enrique Tom\'as Mart\'inez Beltr\'an, Alberto Huertas Celdr\'an
llms
More like this →
arXiv Machine Learning
4d ago

Capturing Membrane Dynamics and Spike Timing of Human Neurons at Scale using Neural Operators

The paper introduces a neural-operator framework that learns a shared operator across multiple neurons of the same cell type, conditioned on biologically interpretable electrophysiological features. This approach scales biophysically realistic neuron models to large populations while preserving rich cellular dynamics and within-cell-type variability. Evaluation shows accurate reproduction of subthreshold membrane dynamics, spike waveforms, firing-rate responses, and spike timing across major cortical inhibitory neuron types, with predicted electrophysiological feature distributions closely matching detailed simulations and human recordings.

By Luca Ghafourpour, Cindy Zhao, Valentin Duruisseaux, Bahareh Tolooshams, Philip H. Wong, Costas A. Anastassiou, Anima Anandkumar
More like this →