Cover Image

The Local Interaction Basis: Identifying Computationally-Relevant and Sparsely Interacting Features in Neural Networks

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Bushnaq, Lucius, Heimersheim, Stefan, Goldowsky-Dill, Nicholas, Braun, Dan, Mendel, Jake, Hänni, Kaarel, Griffin, Avery, Stöhler, Jörn, Wache, Magdalena, Hobbhahn, Marius
Format:	Preprint
Published:	2024
Subjects:	Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Using Degeneracy in the Loss Landscape for Mechanistic Interpretability
by: Bushnaq, Lucius, et al.
Published: (2024)

Interpretability in Parameter Space: Minimizing Mechanistic Description Length with Attribution-based Parameter Decomposition
by: Braun, Dan, et al.
Published: (2025)

Detecting Strategic Deception Using Linear Probes
by: Goldowsky-Dill, Nicholas, et al.
Published: (2025)

Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
by: Braun, Dan, et al.
Published: (2024)

Mathematical Models of Computation in Superposition
by: Hänni, Kaarel, et al.
Published: (2024)

Stochastic Parameter Decomposition
by: Bushnaq, Lucius, et al.
Published: (2025)

Identifying Sparsely Active Circuits Through Local Loss Landscape Decomposition
by: Chrisman, Brianna, et al.
Published: (2025)

Towards evaluations-based safety cases for AI scheming
by: Balesni, Mikita, et al.
Published: (2024)

AI Behind Closed Doors: a Primer on The Governance of Internal Deployment
by: Stix, Charlotte, et al.
Published: (2025)

From Memorization to Reasoning in the Spectrum of Loss Curvature
by: Merullo, Jack, et al.
Published: (2025)

Efficient all-electron Bethe-Salpeter implementation using crystal symmetries
by: Stöhler, Jörn, et al.
Published: (2026)

Disrupted Idylls
by: Stohler, Ursula
Published: (2019)

Cluster-norm for Unsupervised Probing of Knowledge
by: Laurito, Walter, et al.
Published: (2024)

Large Language Models can Strategically Deceive their Users when Put Under Pressure
by: Scheurer, Jérémy, et al.
Published: (2023)

A Principled Basis for Nonequilibrium Network Flows
by: Yang, Ying-Jen, et al.
Published: (2024)

You can remove GPT2's LayerNorm by fine-tuning
by: Heimersheim, Stefan
Published: (2024)

Analogical Reasoning Within a Conceptual Hyperspace
by: Goldowsky, Howard, et al.
Published: (2024)

Technical Report: Evaluating Goal Drift in Language Model Agents
by: Arike, Rauno, et al.
Published: (2025)

Forecasting Frontier Language Model Agent Capabilities
by: Pimpale, Govind, et al.
Published: (2025)

Conditional plasticity of the unit ball of the $\ell_\infty$-sum of finitely many strictly convex Banach spaces
by: Kurik, Kaarel August
Published: (2024)

The Societal Response to Potentially Sentient AI
by: Caviola, Lucius
Published: (2025)

Attribution, Responsibility and Jurisdiction in International Human Rights Law
by: Lucius Caflisch
Published: (2017)

Por um fazer marginal: a questão marioandradina e a fatura oswaldiana da poética de Cacaso
by: Lucius Provase
Published: (2019)

SCALAR: Benchmarking SAE Interaction Sparsity in Toy LLMs
by: Fillingham, Sean P., et al.
Published: (2025)

Ingeborg Weber-Kellermann, Brauch. Familie. Arbeitsleben. Marburg, 1978
by: Hanni Markel
Published: (1985)

Kurt Ranke (ed.), Enzyklopädie des Märchens. Handwörterbuch zur historischen und vergleichenden Erzählforschung. Bd. I–IV, Berlin/New York, 1975–1985
by: Hanni Markel
Published: (1987)

Identitate și etnicitate
by: Hanni Markel
Published: (1994)

Cercetarea etnografică a naționalităților
by: Hanni Markel
Published: (1994)

Landlerii — o enclavă lingvistică, în altă enclavă
by: Hanni Markel
Published: (1996)

Verfassungsstruktur des Judikativen Rechts
by: Hänni, Julia
Published: (2024)

Soziale Arbeit im Kontext Schule
by: Chiapparini, Emanuela, et al.
Published: (2026)

Soziale Arbeit im Kontext Schule
by: Chiapparini, Emanuela, et al.
Published: (2019)

Open Problems in Mechanistic Interpretability
by: Sharkey, Lee, et al.
Published: (2025)

Exploring Sidewalk Sheds in New York City through Chatbot Surveys and Human Computer Interaction
by: Li, Junyi, et al.
Published: (2026)

Large Language Models Often Know When They Are Being Evaluated
by: Needham, Joe, et al.
Published: (2025)

Analyzing Probabilistic Methods for Evaluating Agent Capabilities
by: Højmark, Axel, et al.
Published: (2024)

How to use and interpret activation patching
by: Heimersheim, Stefan, et al.
Published: (2024)

Socio-political implications of the fight against alcoholism and tuberculosis in Colombia, 1910-1925
by: Hanni Jalil-Paier
Published: (2010)

Challenges in the management of chronic noncommunicable diseases by Indonesian community pharmacists
by: Hanni P. PUSPITASARI
Published: (2015)

Presentación
by: Hanni Jalil-Paier
Published: (2018)