paper-with-me

Papers

A Practical Review of Mechanistic Interpretability for Transformer-Based Language Models

2024-07-02 · Daking Rai, Yilun Zhou, Shi Feng, Abulhair Saparov, Ziyu Yao

Mechanistic interpretability (MI) is an emerging sub-field of interpretability that seeks to understand a neural network model by reverse-engineering its internal computations. Recently, MI has garnered significant attention for interpreting transformer-based language models (LMs), resulting in many novel insights yet introducing new challenges. However, there has not been work that comprehensively reviews these insights and challenges, particularly as a guide for newcomers to this field. To fill this gap, we present a comprehensive survey outlining fundamental objects of study in MI, techniques that have been used for its investigation, approaches for evaluating MI results, and significant findings and applications stemming from the use of MI to understand LMs. In particular, we present a roadmap for beginners to navigate the field and leverage MI for their benefit. Finally, we also identify current gaps in the field and discuss potential future directions.

📄 PDF Abstract BibTeX arXiv:2407.02646

Code (1)

dakingrai/awesome-mechanistic-interpretability-lm-papers 공식 구현 pytorch

Tasks

Navigate

Methods 이 논문이 사용한 방법론

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Attention 설명 없음

Similar Papers 제목 키워드 기반

You can remove GPT2's LayerNorm by fine-tuning

2024-09-06 · Stefan Heimersheim

The LayerNorm (LN) layer in GPT-style transformer models has long been a hindrance to mechanistic interpretability. LN is a crucial component required to stabilize the training of large language models, and LN or the sim…

HellaSwag

Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language Models

2026-01-20 · Hengyuan Zhang, Zhihao Zhang, Mingyang Wang, Zunhai Su 외 arxiv

Mechanistic Interpretability (MI) has emerged as a vital approach to demystify the opaque decision-making of Large Language Models (LLMs). However, existing reviews primarily treat MI as an observational science, summari…

Open Problems in Mechanistic Interpretability

2025-01-27 · Lee Sharkey, Bilal Chughtai, Joshua Batson, Jack Lindsey 외

Mechanistic interpretability aims to understand the computational mechanisms underlying neural networks' capabilities in order to accomplish concrete scientific and engineering goals. Progress in this field thus promises…

Interpreting Transformers Through Attention Head Intervention

2026-01-07 · Mason Kadem, Rong Zheng arxiv

Neural networks are growing more capable on their own, but we do not understand their neural mechanisms. Understanding these mechanisms' decision-making processes, or mechanistic interpretability, enables (1) accountabil…

Mechanistic Interpretability for AI Safety -- A Review

2024-04-22 · Leonard Bereska, Efstratios Gavves

Understanding AI systems' inner workings is critical for ensuring value alignment and safety. This review explores mechanistic interpretability: reverse engineering the computational mechanisms and representations learne…