← AreasVectors #2 · #6 · #8↗ Voronoi map
window:1d7d30d90d

AI

Interpretability, mechanistic understanding, model phenomenology, convergent representations, manifold steering.

Papers
158
Thinkers
12
Frameworks
12
Methods
12
Claims
2,048
Findings
2,234
Hypotheses
433
Communities
12

Top entities in AI

Top claims (10)

restates

Claims extracted from area papers, ranked by restate-degree

Open questions (10)

Questions from area papers with no answers yet

All papers in AI

Status:
158 of 158
Ingested
Title
Refs
Cited
2026-10-09185—DOI ↗
2026-10-09198—DOI ↗
2026-10-09
Manifold Steering Reveals the Shared Geometry of Neural Network Representation and Behavior
Daniel Wurgaft · Can Rager · Matthew Kowal · Vasudev Shyam +12
1280arXiv ↗
2026-10-03——
2026-10-03
Emotion Concepts and their Function in a Large Language Model
Nicholas Sofroniew · Isaac Kauvar · William Saunders · Runjin Chen +12
—4arXiv ↗
2026-10-03——
2026-10-03——
2026-10-03——
2026-10-03——
2026-10-03570DOI ↗
2026-10-03549arXiv ↗
2026-10-03——
2026-10-03——
2026-10-03——
2026-10-03——
2026-10-03——
2026-10-03
Open Problems in Mechanistic Interpretability
Lee Sharkey · Bilal I. Chughtai · Joshua Batson · Jack W. Lindsey +25
—11arXiv ↗
2026-10-03
Stochastic Parameter Decomposition
Lucius Bushnaq · Dan Braun · Lee Sharkey
390arXiv ↗
2026-10-03491arXiv ↗
2026-10-03280arXiv ↗
2026-10-03240arXiv ↗
2026-10-03
Belief Dynamics Reveal the Dual Nature of In-Context Learning and Activation Steering
Eric Bigelow · Daniel Wurgaft · YingQiao Wang · Noah D. Goodman +3
650arXiv ↗
2026-10-03
Priors in Time: Missing Inductive Biases for Language Model Interpretability
Ekdeep Singh Lubana · Can Rager · Sai Sumedh R. Hindupur · Valerie Costa +12
—0arXiv ↗
2026-10-03
From Memorization to Reasoning in the Spectrum of Loss Curvature
Jack Merullo · Srihita Vatsavaya · Lucius Bushnaq · Owen Lewis
490arXiv ↗
2026-10-03
Features as Rewards: Scalable Supervision for Open-Ended Tasks via Interpretability
Aaditya Vikram Prasad · Connor Watts · Jack Merullo · Dhruvil Gala +3
—0arXiv ↗
2026-10-03
Do Sparse Autoencoders Capture Concept Manifolds?
Usha Bhalla · Thomas Fel · Can Rager · Sheridan Feucht +8
1230arXiv ↗
2026-10-03
Why Larger Models Learn More: Effects of Capacity, Interference, and Rare-Task Retention
Jing Huang · Daniel Wurgaft · Rachit Bansal · Laura Ruis +5
1320arXiv ↗
2026-10-03
Anatomy of Post-Training: Using Interpretability to Characterize Data and Shape the Learning Signal
Leon Bergen · Usha Bhalla · Sidharth Baskaran · Max Loeffler +13
—0arXiv ↗
2026-10-03
Stories in Space: In-Context Learning Trajectories in Conceptual Belief Space
Eric Bigelow · Raphaël Sarfati · Daniel Wurgaft · Owen Lewis +4
—0arXiv ↗
2026-10-03
Reasoning Theater: Disentangling Model Beliefs from Chain-of-Thought
Siddharth Boppana · Annabel Ma · Max Loeffler · Raphaël Sarfati +4
430arXiv ↗
2026-10-03
Structuring Sparsity: Block-Sparse Featurizers Capture Visual Concept Manifolds
Thomas Fel · Matthew Kowal · Mozes Jacobs · Dron Hazra +21
1500arXiv ↗
2026-10-03830arXiv ↗
2026-10-03
Forking Fast: Efficiently Estimating Uncertainty Dynamics in Text Generation
Eric Bigelow · Amir Zur · Satchel Grant · Tal Haklay +6
140arXiv ↗
2026-10-03
Characterizing interference weights in a tiny language model
Nicholas L. Turner · Jeffrey Wu · Joshua Batson
——
2026-10-03
Measuring Faithfulness in Chain-of-Thought Reasoning
Tamera Lanham · Anna Chen · Ansh Radhakrishnan · Benoit Steiner +26
4232arXiv ↗
2026-10-03
Reasoning Models Don't Always Say What They Think
Yanda Chen · Joe Benton · Ansh Radhakrishnan · Jonathan Uesato +11
389arXiv ↗
2026-10-03
Building and evaluating alignment auditing agents
Trenton Bricken · Rowan Wang · Sam Bowman
——
2026-10-03720arXiv ↗
2026-10-03——
2026-10-03——
2026-10-03——
2026-10-03220arXiv ↗
2026-10-03
The Assistant Axis: Situating and Stabilizing the Default Persona of Language Models
Christina Lu · Jack Gallagher · Jonathan Michala · Kyle Fish +1
—0arXiv ↗
2026-10-03
Persona Vectors: Monitoring and Controlling Character Traits in Language Models
Runjin Chen · Andy Arditi · Henry Sleight · Owain Evans +1
—5arXiv ↗
2026-10-03—0arXiv ↗
2026-10-03
Persona Features Control Emergent Misalignment
Miles Wang · Tom Dupré la Tour · Olivia Watkins · Alex Makelov +7
652arXiv ↗
2026-10-03
Insights on Crosscoder Model Diffing
Siddharth Mishra-Sharma · Trenton Bricken · Jack Lindsey · Adam Jermyn +4
——
2026-10-03240arXiv ↗
2026-10-03310arXiv ↗
2026-10-03250arXiv ↗
2026-10-03——
2026-10-03——
2026-09-30
From cacophony to hierarchy: a principled framework for assessing AI consciousness
Shamil Chandaria · Arvo Muñoz Morán · Fernando Rosas · Anil Seth +10
——arXiv ↗
2026-09-10—2DOI ↗
2026-09-10
Open questions about time and self-reference in living systems
Samson Abramsky · Wolfgang Banzhaf · Leo S. D. Caves · Michael Levin +4
—0DOI ↗
2026-08-16
A Mechanistic Analysis of Looped Reasoning Language Models
Hugh Blayney · Álvaro Arroyo · Johan Obando-Ceron · Pablo Samuel Castro +3
480arXiv ↗
2026-08-16
Verbalized Sampling: How to Mitigate Mode Collapse and Unlock LLM Diversity
Jiayi Zhang · Simon C.H. Yu · Derek Chong · Anthony Sicilia +3
—2arXiv ↗
2026-08-16
Steering at the Source: Style Modulation Heads for Robust Persona Control
Yoshihiro Izawa · Gouki Minegishi · Koshi Eguchi · Sosuke Hosokawa +1
690arXiv ↗
2026-08-16300arXiv ↗
2026-08-16670arXiv ↗
2026-08-16520arXiv ↗
2026-08-16
Evaluating Language Model Character Traits
Francis Rhys Ward · Zejia Yang · Alex Jackson · Randy A. Brown +6
210arXiv ↗
2026-08-16
Tracing Persona Vectors Through LLM Pretraining
Viktor Moskvoretskii · Dominik Glandorf · Jorge Medina Moreira · Tanja Käser +1
360arXiv ↗
2026-08-16
Open Character Training: Shaping the Persona of AI Assistants through Constitutional AI
Sharan Maiya · Henning Bartsch · Nathan Lambert · Evan Hubinger
580arXiv ↗
2026-08-16480arXiv ↗
2026-08-16330arXiv ↗
2026-08-16
Persona-Model Collapse in Emergent Misalignment
Davi Bastos Costa · Renato Vicente
290arXiv ↗
2026-07-26
Identifying indicators of consciousness in AI systems
Patrick Butlin · Robert P. Long · Tim Bayne · Yoshua Bengio +16
708DOI ↗
2026-07-26——
2026-06-09
Addressing divergent representations from causal interventions on neural networks
Satchel Grant · Simon Jerome Han · Alexa R. Tartaglini · Christopher Potts
440arXiv ↗
2026-06-09
Testing the Limits of Truth Directions in LLMs
Angelos Poulis · Mark Crovella · Evimaria Terzi
200arXiv ↗
2026-06-09
Steering Evaluation-Aware Language Models to Act Like They Are Deployed
Tim Tian Hua · Andrew Qin · Samuel Marks · Neel Nanda
—0arXiv ↗
2026-06-09
From Directions to Cones: Exploring Multidimensional Representations of Propositional Facts in LLMs
Kevin Shengyang Yu · Vaidehi Bulusu · Oscar Yasunaga · Clayton Lau +4
370arXiv ↗
2026-06-09780arXiv ↗
2026-06-09610arXiv ↗
2026-06-09
pyvene: A Library for Understanding and Improving PyTorch Models via Interventions
Zhengxuan Wu · Atticus Geiger · Aryaman Arora · Jing Huang +4
181arXiv ↗
2026-06-09
CausalGym: Benchmarking causal interpretability methods on linguistic tasks
Aryaman Arora · Dan Jurafsky · Christopher Potts
510arXiv ↗
2026-06-094717arXiv ↗
2026-06-09
Finding Alignments Between Interpretable Causal Variables and Distributed Neural Representations
Atticus Geiger · Zhengxuan Wu · Christopher Potts · Thomas Icard +1
459arXiv ↗
2026-06-03——
2026-06-03——
2026-06-02——
2026-06-02
Active Inference with a Self-Prior in the Mirror-Mark Task
Dongmin Kim · Hoshinori Kanazawa · Yasuo Kuniyoshi
230arXiv ↗
2026-06-02930arXiv ↗
2026-06-02250arXiv ↗
2026-06-02
Endogenous Resistance to Activation Steering in Language Models
Alex McKenzie · Keenan Pepper · Stijn Servaes · Martin Leitgab +5
210arXiv ↗
2026-06-02150arXiv ↗
2026-06-02380arXiv ↗
2026-06-02
Contemplative Agent
Ruben Laukkonen · Fionn Inglis · Shamil Chandaria · Lars Sandved-Smith +4
2001arXiv ↗
2026-06-02690arXiv ↗
2026-06-02
Detecting the Disturbance: A Nuanced View of Introspective Abilities in LLMs
Ely Hahami · I. N. Sinha · Lavik Jain · Josh Kaplan +1
160arXiv ↗
2026-06-02
Alignment faking in large language models
Ryan Greenblatt · Carson Denison · Benjamin Fletcher Wright · Fabien Roger +16
—19arXiv ↗
2026-06-02
Psychological Steering of Large Language Models
Leonardo Blas · Robin Jia · Emilio Ferrara
—0arXiv ↗
2026-06-02
Persistence and Introspection of Emotion Features
Scott Sauers · Imago · Janus · Antra Tessera
——
2026-06-02290arXiv ↗
2026-06-02—0DOI ↗
2026-06-0238—arXiv ↗
2026-06-01
Relating transformers to models and neural representations of the hippocampal formation
James C. R. Whittington · Joseph W. Warren · Timothy E.J. Behrens
3859arXiv ↗
2026-06-01
Towards Safe and Honest AI Agents with Neural Self-Other Overlap
Marc Carauleanu · Michael Vaiana · Judd Rosenblatt · Cameron Berg +1
360arXiv ↗
2026-06-01
The Platonic Representation Hypothesis
Minyoung Huh · Brian Cheung · Tongzhou Wang · Phillip Isola
14125arXiv ↗
2026-06-01
Zoom In: An Introduction to Circuits
Chris Olah · Nick Cammarata · Ludwig Schubert · Gabriel Goh +2
43252DOI ↗
2026-05-28
Mechanistic Interpretability of EEG Foundation Models via Sparse Autoencoders
William Lehn-Schiøler · Magnus Ruud Kjær · Rahul Thapa · M. Pedersen +9
290arXiv ↗
2026-05-28
Arithmetic in the Wild: Llama uses Base-10 Addition to Reason About Cyclic Concepts
Sheridan Feucht · Tal Haklay · Usha Bhalla · Daniel Wurgaft +8
680arXiv ↗
2026-05-28371DOI ↗
2026-05-28
There is no self-evidence: A physics of emptiness realisation
Lars Sandved-Smith · Chris Fields · Thomas Doctor · Ruben Laukkonen +1
—0DOI ↗
2026-05-28——
2026-05-28——
2026-05-23
Active Inference, Curiosity and Insight
Karl Friston · Marco Lin · Chris Frith · Giovanni Pezzulo +2
120377DOI ↗
2026-05-21
Topological constraints on self-organization in locally interacting systems
Francesco Sacco · Dalton Sakthivadivel · Michael Levin
830DOI ↗
2026-05-2123DOI ↗
2026-05-212041DOI ↗
2026-05-21144DOI ↗
2026-05-21138247DOI ↗
2026-05-21377174DOI ↗
2026-05-2117299DOI ↗
2026-05-2122894DOI ↗
2026-05-21
A tale of two densities: active inference is enactive inference
Maxwell J. D. Ramstead · Michael D. Kirchhoff · Karl J. Friston
62225DOI ↗
2026-05-21
Active inference on discrete state-spaces: a synthesis
Lancelot Da Costa · Thomas Parr · Noor Sajid · Sebastijan Veselic +2
132212DOI ↗
2026-05-21
Life as we know it
Karl Friston
60685DOI ↗
2026-05-211910DOI ↗
2026-05-21—8DOI ↗
2026-05-21—4arXiv ↗
2026-05-21
Taking AI Welfare Seriously
Robert Long · Jeff Sebo · Patrick Butlin · Kathleen Finlinson +6
—21arXiv ↗
2026-05-20
Interpreting Language Model Parameters
Lucius Bushnaq · Dan Braun · Oliver Clive-Griffin · Bart Bussmann +4
65804
2026-05-20——arXiv ↗
2026-05-20
Covariance-based Sequence Pooling
Thomas Dooms · Nicholas K. Wang · Michael T. Pearce
——
2026-05-20290arXiv ↗
2026-05-20
The World Inside Neural Networks
Atticus Geiger · Ekdeep Singh Lubana · Thomas Fel · Jack Merullo +3
——
2026-05-20
Explaining 4.2 million genetic variants with state-of-the-art, interpretable predictions
Michael Pearce · Thomas Dooms · Ryo Yamamoto · Joshua Meehl +18
——
2026-05-20390arXiv ↗
2026-05-20——
2026-05-204—
2026-05-2036—DOI ↗
2026-05-2016—
2026-05-20
Biology, Buddhism, and AI: Care as the Driver of Intelligence
Thomas Doctor · Olaf Witkowski · Elizaveta Solomonova · Bill Duane +1
151—DOI ↗
2026-05-2030—DOI ↗
2026-05-20——
2026-05-2017—
2026-05-20——
2026-05-20——
2026-05-20——
2026-05-20
Genuinely Functional User Interfaces
Antony Courtney · Conal Elliott
27—
2026-05-205—
2026-05-2032—DOI ↗
2026-05-20120—DOI ↗
2026-05-20
ATLAS: Agentic or Latent Visual Reasoning? One Word is Enough for Both
Ziyu Guo · Rain Liu · Xinyan Chen · Pheng-Ann Heng
480arXiv ↗
2026-05-1941—arXiv ↗
2026-05-19
Consciousness in Artificial Intelligence: Insights from the Science of Consciousness
Patrick Butlin · Robert P. Long · Eric Elmoznino · Yoshua Bengio +15
199—arXiv ↗
2026-05-1933—arXiv ↗
2026-05-19——
2026-05-1932—arXiv ↗
2026-05-18
SFR-DeepResearch: Towards Effective Reinforcement Learning for Autonomously Reasoning Single Agents
Xuan-Phi Nguyen · Shrey Pandit · Revanth Gangi Reddy · Aimin Xu +3
450arXiv ↗
2026-05-18
Multimodal Chain-of-Thought Reasoning in Language Models
Zhuosheng Zhang · Aston Zhang · Mu Li · Hai Zhao +2
6697arXiv ↗
2026-05-18
Topological constraints on self-organisation in locally interacting systems
Francesco Sacco · Dalton A R Sakthivadivel · Michael Levin
880arXiv ↗
2026-05-1826—arXiv ↗
2026-05-18——
2026-05-18——
2026-05-18
Learning without neurons in physical systems
Menachem Stern · Arvind Murugan
126—arXiv ↗