Imagen de Portada

Inference-Time Decomposition of Activations (ITDA): A Scalable Approach to Interpreting Large Language Models

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Leask, Patrick, Nanda, Neel, Moubayed, Noura Al
Formato:	Preprint
Publicado:	2025
Materias:	Machine Learning
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Sparse Autoencoders Do Not Find Canonical Units of Analysis
por: Leask, Patrick, et al.
Publicado: (2025)

BatchTopK Sparse Autoencoders
por: Bussmann, Bart, et al.
Publicado: (2024)

Breaking Down Financial News Impact: A Novel AI Approach with Geometric Hypergraphs
por: Harit, Anoushka, et al.
Publicado: (2024)

Towards Best Practices of Activation Patching in Language Models: Metrics and Methods
por: Zhang, Fred, et al.
Publicado: (2023)

Towards Principled Evaluations of Sparse Autoencoders for Interpretability and Control
por: Makelov, Aleksandar, et al.
Publicado: (2024)

Explorations of Self-Repair in Language Models
por: Rushing, Cody, et al.
Publicado: (2024)

Transcoders Find Interpretable LLM Feature Circuits
por: Dunefsky, Jacob, et al.
Publicado: (2024)

Controllable Image Generation with Composed Parallel Token Prediction
por: Stirling, Jamie, et al.
Publicado: (2026)

Investigating Permutation-Invariant Discrete Representation Learning for Spatially Aligned Images
por: Stirling, Jamie S. J., et al.
Publicado: (2026)

Video Prediction of Dynamic Physical Simulations With Pixel-Space Spatiotemporal Transformers
por: Slack, Dean L, et al.
Publicado: (2025)

Disentangling Racial Phenotypes: Fine-Grained Control of Race-related Facial Phenotype Characteristics
por: Yucer, Seyma, et al.
Publicado: (2024)

InverseScope: Scalable Activation Inversion for Interpreting Large Language Models
por: Luo, Yifan, et al.
Publicado: (2025)

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

Controllable Image Generation with Composed Parallel Token Prediction
por: Stirling, Jamie, et al.
Publicado: (2024)

Interpreting Attention Layer Outputs with Sparse Autoencoders
por: Kissane, Connor, et al.
Publicado: (2024)

Interpretable Embeddings with Sparse Autoencoders: A Data Analysis Toolkit
por: Jiang, Nick, et al.
Publicado: (2025)

RelP: Faithful and Efficient Circuit Discovery in Language Models via Relevance Patching
por: Jafari, Farnoush Rezaei, et al.
Publicado: (2025)

Thought Branches: Interpreting LLM Reasoning Requires Resampling
por: Macar, Uzay, et al.
Publicado: (2025)

Do I Know This Entity? Knowledge Awareness and Hallucinations in Language Models
por: Ferrando, Javier, et al.
Publicado: (2024)

Difficulties with Evaluating a Deception Detector for AIs
por: Smith, Lewis, et al.
Publicado: (2025)

Large Language Model Compression via the Nested Activation-Aware Decomposition
por: Lu, Jun, et al.
Publicado: (2025)

Reasoning-Finetuning Repurposes Latent Representations in Base Models
por: Ward, Jake, et al.
Publicado: (2025)

Understanding Reasoning in Thinking Language Models via Steering Vectors
por: Venhoff, Constantin, et al.
Publicado: (2025)

Overcoming Sparsity Artifacts in Crosscoders to Interpret Chat-Tuning
por: Minder, Julian, et al.
Publicado: (2025)

An Interpretable and Scalable Framework for Evaluating Large Language Models
por: Qu, Xinhao, et al.
Publicado: (2026)

Eliciting Secret Knowledge from Language Models
por: Cywiński, Bartosz, et al.
Publicado: (2025)

Summing Up the Facts: Additive Mechanisms Behind Factual Recall in LLMs
por: Chughtai, Bilal, et al.
Publicado: (2024)

LLMs for LLMs: A Structured Prompting Methodology for Long Legal Documents
por: Klem, Strahinja, et al.
Publicado: (2025)

Fine-Grained Uncertainty Decomposition in Large Language Models: A Spectral Approach
por: Walha, Nassim, et al.
Publicado: (2025)

Towards eliciting latent knowledge from LLMs with mechanistic interpretability
por: Cywiński, Bartosz, et al.
Publicado: (2025)

Inference-Time Reward Hacking in Large Language Models
por: Khalaf, Hadi, et al.
Publicado: (2025)

SAEBench: A Comprehensive Benchmark for Sparse Autoencoders in Language Model Interpretability
por: Karvonen, Adam, et al.
Publicado: (2025)

Everything is a Video: Unifying Modalities through Next-Frame Prediction
por: Hudson, G. Thomas, et al.
Publicado: (2024)

Base Models Know How to Reason, Thinking Models Learn When
por: Venhoff, Constantin, et al.
Publicado: (2025)

Model Organisms for Emergent Misalignment
por: Turner, Edward, et al.
Publicado: (2025)

Mitigating Over-Refusal in Aligned Large Language Models via Inference-Time Activation Energy
por: Jiang, Eric Hanchen, et al.
Publicado: (2025)

Extracting Interpretable Task-Specific Circuits from Large Language Models for Faster Inference
por: García-Carrasco, Jorge, et al.
Publicado: (2024)

StructTransform: A Scalable Attack Surface for Safety-Aligned Large Language Models
por: Yoosuf, Shehel, et al.
Publicado: (2025)

Proof of Concept: Multi-Target Wildfire Risk Prediction and Large Language Model Synthesis
por: Caron, Nicolas, et al.
Publicado: (2026)

Simple LLM Baselines are Competitive for Model Diffing
por: Kempf, Elias, et al.
Publicado: (2026)