v1v2 (latest)

Towards Best Practices of Activation Patching in Language Models: Metrics and Methods

International Conference on Learning Representations (ICLR), 2023

27 September 2023

Fred Zhang

Neel Nanda

LLMSV

ArXiv (abs)PDF HTML HuggingFace (4 upvotes)

Papers citing "Towards Best Practices of Activation Patching in Language Models: Metrics and Methods"

50 / 127 papers shown

Title
No Free Lunch in Language Model Bias Mitigation? Targeted Bias Reduction Can Exacerbate Unmitigated LLM Biases Shireen Chand Faith Baca Emilio Ferrara 104 0 0 23 Nov 2025
Understanding Counting Mechanisms in Large Language and Vision-Language Models Hosein Hasani Amirmohammad Izadi Fatemeh Askari Mobin Bagherian Sadegh Mohammadian Mohammad Izadi M. Baghshah 68 0 0 21 Nov 2025
Anatomy of an Idiom: Tracing Non-Compositionality in Language Models Andrew Gomes 166 0 0 20 Nov 2025
BlockCert: Certified Blockwise Extraction of Transformer Mechanisms Sandro Andric 56 0 0 20 Nov 2025
Training Language Models to Explain Their Own Computations Belinda Z. Li Zifan Carl Guo Vincent Huang Jacob Steinhardt Jacob Andreas LRM 160 3 0 11 Nov 2025
APP: Accelerated Path Patching with Task-Specific Pruning Frauke Andersen William Rudman Ruochen Zhang Carsten Eickhoff 52 0 0 07 Nov 2025
Addressing divergent representations from causal interventions on neural networks Satchel Grant Simon Jerome Han Alexa R. Tartaglini Christopher Potts CML 433 0 0 06 Nov 2025
LLMs Process Lists With General Filter Heads Arnab Sen Sharma Giordano Rogers Natalie Shapira David Bau 120 0 0 30 Oct 2025
Explainability of Large Language Models: Opportunities and Challenges toward Generating Trustworthy Explanations Shahin Atakishiyev H. Babiker Jiayi Dai Nawshad Farruque Teruaki Hayashi ... Md Abed Rahman Iain Smith Mi-Young Kim Osmar R. Zaïane Randy Goebel LRM 137 0 0 20 Oct 2025
How role-play shapes relevance judgment in zero-shot LLM rankers Yumeng Wang Jirui Qi Catherine Chen Panagiotis Eustratiadis Suzan Verberne 66 0 0 20 Oct 2025
DARTS-GT: Differentiable Architecture Search for Graph Transformers with Quantifiable Instance-Specific Interpretability Analysis Shruti Sarika Chakraborty Peter Minary 143 0 0 16 Oct 2025
Analysing Moral Bias in Finetuned LLMs through Mechanistic Interpretability Bianca Raimondi Daniela Dalbagno Maurizio Gabbrielli AI4CE 49 0 0 14 Oct 2025
The Curious Case of Factual (Mis)Alignment between LLMs' Short- and Long-Form Answers Saad Obaid ul Islam Anne Lauscher Goran Glavaš HILM 174 0 0 13 Oct 2025
Medical Interpretability and Knowledge Maps of Large Language Models Razvan Marinescu Victoria-Elisabeth Gruber Diego Fajardo FAtt AI4MH 202 0 0 13 Oct 2025
Discursive Circuits: How Do Language Models Understand Discourse Relations? Yisong Miao Min-Yen Kan 136 2 0 13 Oct 2025
$Causality $\neq$ Decodability, and Vice Versa: Lessons from Interpreting Counting ViTs$ Causality $\neq$ Decodability, and Vice Versa: Lessons from Interpreting Counting ViTs Lianghuan Huang Yingshan Chang CML 56 0 0 10 Oct 2025
Inverse-Free Wilson Loops for Transformers: A Practical Diagnostic for Invariance and Order Sensitivity Edward Y. Chang Ethan Chang 80 1 0 09 Oct 2025
Validation of Various Normalization Methods for Brain Tumor Segmentation: Can Federated Learning Overcome This Heterogeneity? Jan Fiszer Dominika Ciupek Maciej Malawski FedML 196 0 0 08 Oct 2025
Reproducing and Extending Causal Insights Into Term Frequency Computation in Neural Rankers Cile van Marken Roxana Petcu CML 160 0 0 08 Oct 2025
Mechanistic Interpretability as Statistical Estimation: A Variance Analysis of EAP-IG Maxime Méloux François Portet Maxime Peyrard 155 1 0 01 Oct 2025
Query Circuits: Explaining How Language Models Answer User Prompts Tung-Yu Wu Fazl Barez ReLM LRM 121 0 0 29 Sep 2025
Toward Preference-aligned Large Language Models via Residual-based Model Steering Lucio La Cava Andrea Tagarelli LLMSV 144 0 0 28 Sep 2025
What Matters More For In-Context Learning under Matched Compute Budgets: Pretraining on Natural Text or Incorporating Targeted Synthetic Examples? Mohammed Sabry Anya Belz 79 0 0 26 Sep 2025
Can Large Language Models Develop Gambling Addiction? Seungpil Lee Donghyeon Shin Yunjeong Lee Sundong Kim AIFin AI4CE 154 1 0 26 Sep 2025
How Persuasive is Your Context? Tu Nguyen Kevin Du Alexander Miserlis Hoyle Ryan Cotterell 96 0 0 22 Sep 2025
V-SEAM: Visual Semantic Editing and Attention Modulating for Causal Interpretability of Vision-Language Models Qidong Wang Junjie Hu Ming Jiang 72 0 0 18 Sep 2025
Statistical Methods in Generative AI Edgar Dobriban 285 3 0 08 Sep 2025
A Review of Developmental Interpretability in Large Language Models Ihor Kendiukhov ELM 140 0 0 19 Aug 2025
How Causal Abstraction Underpins Computational Explanation Atticus Geiger Jacqueline Harding Thomas Icard 105 2 0 15 Aug 2025
Understanding and Mitigating Political Stance Cross-topic Generalization in Large Language Models J. Zhang Shu Yang Junchao Wu Yang Li Haiyan Zhao 184 1 0 04 Aug 2025
Unveiling the Influence of Amplifying Language-Specific Neurons Inaya Rahmanisa Lyzander Marciano Andrylie Mahardika Krisna Ihsani Alfan Farizki Wicaksono Haryo Akbarianto Wibowo Alham Fikri Aji 122 0 0 30 Jul 2025
Dissecting Persona-Driven Reasoning in Language Models via Activation Patching Ansh Poonia Maeghal Jain 177 0 0 28 Jul 2025
Latent Concept Disentanglement in Transformer-based Language Models Guan Zhe Hong Bhavya Vasudeva Willie Neiswanger Cyrus Rashtchian Prabhakar Raghavan Rina Panigrahy ReLM LRM 315 2 0 20 Jun 2025
Rethinking Explainability in the Era of Multimodal AI Chirag Agarwal 195 1 0 16 Jun 2025
Universal Jailbreak Suffixes Are Strong Attention Hijackers Matan Ben-Tov Mor Geva Mahmood Sharif 152 0 0 15 Jun 2025
Path Channels and Plan Extension Kernels: a Mechanistic Description of Planning in a Sokoban RNN Mohammad Taufeeque Aaron David Tucker Adam Gleave Adrià Garriga-Alonso 246 0 0 11 Jun 2025
Learning Distribution-Wise Control in Representation Space for Language Models Chunyuan Deng Ruidi Chang Hanjie Chen 239 2 0 07 Jun 2025
Establishing Trustworthy LLM Evaluation via Shortcut Neuron AnalysisAnnual Meeting of the Association for Computational Linguistics (ACL), 2025 Kejian Zhu Shangqing Tu Zhuoran Jin Lei Hou Juanzi Li Jun Zhao KELM 195 0 0 04 Jun 2025
Can LLMs Reason Abstractly Over Math Word Problems Without CoT? Disentangling Abstract Formulation From Arithmetic Computation Ziling Cheng Meng Cao Leila Pishdad Yanshuai Cao Jackie Chi Kit Cheung LRM 348 3 0 29 May 2025
An Empirical Study of the Anchoring Effect in LLMs: Existence, Mechanism, and Potential Mitigations Yiming Huang Biquan Bie Zuqiu Na Weilin Ruan Songxin Lei Yutao Yue Xinlei He 211 2 0 21 May 2025
Explaining Neural Networks with Reasons Levin Hornischer Hannes Leitgeb FAtt AAML MILM 287 0 0 20 May 2025
Language Models Are Capable of Metacognitive Monitoring and Control of Their Internal Activations Li Ji-An Hua-Dong Xiong Robert C. Wilson Marcelo G. Mattar M. Benna 287 11 0 19 May 2025
SPIRIT: Patching Speech Language Models against Jailbreak Attacks Amirbek Djanibekov Nurdaulet Mukhituly Kentaro Inui Hanan Aldarmaki Nils Lukas AAML 272 1 0 18 May 2025
Rethinking Circuit Completeness in Language Models: AND, OR, and ADDER Gates Hang Chen Jiaying Zhu Xinyu Yang Wenya Wang LRM 316 3 0 15 May 2025
Interpreting Multilingual and Document-Length Sensitive Relevance Computations in Neural Retrieval Models through Axiomatic Causal InterventionsAnnual International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR), 2025 Oliver Savolainen Dur e Najaf Amjad Roxana Petcu AAML 171 0 0 04 May 2025
Self-Ablating Transformers: More Interpretability, Less Sparsity Jeremias Ferrao Luhan Mikaelson Keenan Pepper Natalia Perez-Campanero Antolin MILM 246 0 0 01 May 2025
Towards Understanding the Nature of Attention with Low-Rank Sparse Decomposition Zhengfu He Jiadong Wang Rui Lin Xuyang Ge Wentao Shu Qiong Tang J.N. Zhang Xipeng Qiu 262 1 0 29 Apr 2025
Functional Abstraction of Knowledge Recall in Large Language Models Zijian Wang Chang Xu KELM 208 1 0 20 Apr 2025
MIB: A Mechanistic Interpretability Benchmark Aaron Mueller Atticus Geiger Sarah Wiegreffe Dana Arad Iván Arcuschin ... Alessandro Stolfo Martin Tutek Amir Zur David Bau Yonatan Belinkov 588 7 0 17 Apr 2025
How do Large Language Models Understand Relevance? A Mechanistic Interpretability Perspective Qi Liu Jiaxin Mao Ji-Rong Wen LRM 246 3 0 10 Apr 2025