Seeing Through Circuits: Faithful Mechanistic Interpretability for Vision Transformers

• URL: http://arxiv.org/abs/2604.14477v1
• Date: Wed, 15 Apr 2026 23:25:46 GMT
• Title: Seeing Through Circuits: Faithful Mechanistic Interpretability for Vision Transformers
• Abstract summary: We investigate whether useful mechanistic circuits can be identified through computational graphs in vision transformers.<n>We propose an effective method for Automatic Visual Circuit Discovery (Vi-CD) that recovers class-specific circuits for classification.<n>We find that insightful and actionable edge-based circuits can be recovered from vision transformers, adding transparency to the internal computations of these models.
• Score: 54.2202951543048
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Transparency of neural networks' internal reasoning is at the heart of interpretability research, adding to trust, safety, and understanding of these models. The field of mechanistic interpretability has recently focused on studying task-specific computational graphs, defined by connections (edges) between model components. Such edge-based circuits have been defined in the context of large language models, yet vision-based approaches so far only consider neuron-based circuits. These tell which information is encoded, but not how it is routed through the complex wiring of a neural network. In this work, we investigate whether useful mechanistic circuits can be identified through computational graphs in vision transformers. We propose an effective method for Automatic Visual Circuit Discovery (Vi-CD) that recovers class-specific circuits for classification, identifies circuits underlying typographic attacks in CLIP, and discovers circuits that lend themselves for steering to correct harmful model behavior. Overall, we find that insightful and actionable edge-based circuits can be recovered from vision transformers, adding transparency to the internal computations of these models.

Related papers

• Mechanistic Interpretability for Neural Networks: Circuits, Sparse Features and Symbolic Reasoning [0.0]
  mechanistic interpretability is an emerging field that seeks to reverse-engineer the internal algorithms of modern neural networks.<n>The paper provides a detailed examination of Transformer circuit analysis, exploring how internal components drive complex tasks and in-context learning.<n>It then tackles the core challenge of superposition and polysemanticity, demonstrating how tools like Sparse Autoencoders (SAEs) and transcoders can decompose tangled network activations into distinct, human-interpretable features.
  arXiv  Detail & Related papers  (2026-07-08T12:04:38Z)
• Scalable Circuit Learning for Interpreting Large Language Models [49.202271807046365]
  We propose CircuitLasso, a scalable circuit-learning approach based on sparse linear regression.<n>For interpretability, CircuitLasso efficiently uncovers relationships among SAE features, showing how human-interpretable semantic features propagate through the model and influence its predictions.
  arXiv  Detail & Related papers  (2026-06-15T16:40:43Z)
• Circuit Compositions: Exploring Modular Structures in Transformer-Based Language Models [22.89563355840371]
  We study the modularity of neural networks by analyzing circuits for highly compositional subtasks within a language model.<n>Our results indicate that functionally similar circuits exhibit both notable node overlap and cross-task faithfulness.
  arXiv  Detail & Related papers  (2024-10-02T11:36:45Z)
• InterpBench: Semi-Synthetic Transformers for Evaluating Mechanistic Interpretability Techniques [8.383131649026804]
  We train simple neural networks using a stricter version of Interchange Intervention Training (IIT) which we call Strict IIT (SIIT)<n>We evaluate SIIT on sparse transformers produced by the Tracr tool and find that SIIT models maintain Tracr's original circuit while being more realistic.
  arXiv  Detail & Related papers  (2024-07-19T17:46:51Z)
• Transformer Circuit Faithfulness Metrics are not Robust [0.04260910081285213]
  We measure circuit 'faithfulness' by ablating portions of the model's computation. We conclude that existing circuit faithfulness scores reflect both the methodological choices of researchers as well as the actual components of the circuit. The ultimate goal of mechanistic interpretability work is to understand neural networks, so we emphasize the need for more clarity in the precise claims being made about circuits.
  arXiv  Detail & Related papers  (2024-07-11T17:59:00Z)
• Automatic Discovery of Visual Circuits [66.99553804855931]
  We explore scalable methods for extracting the subgraph of a vision model's computational graph that underlies recognition of a specific visual concept. We find that our approach extracts circuits that causally affect model output, and that editing these circuits can defend large pretrained models from adversarial attacks.
  arXiv  Detail & Related papers  (2024-04-22T17:00:57Z)
• CktGNN: Circuit Graph Neural Network for Electronic Design Automation [67.29634073660239]
  This paper presents a Circuit Graph Neural Network (CktGNN) that simultaneously automates the circuit topology generation and device sizing. We introduce Open Circuit Benchmark (OCB), an open-sourced dataset that contains $10$K distinct operational amplifiers. Our work paves the way toward a learning-based open-sourced design automation for analog circuits.
  arXiv  Detail & Related papers  (2023-08-31T02:20:25Z)
• Can Transformers Learn to Solve Problems Recursively? [9.5623664764386]
  This paper examines the behavior of neural networks learning algorithms relevant to programs and formal verification. By reconstructing these algorithms, we are able to correctly predict 91 percent of failure cases for one of the approximated functions.
  arXiv  Detail & Related papers  (2023-05-24T04:08:37Z)
• Pretraining Graph Neural Networks for few-shot Analog Circuit Modeling and Design [68.1682448368636]
  We present a supervised pretraining approach to learn circuit representations that can be adapted to new unseen topologies or unseen prediction tasks. To cope with the variable topological structure of different circuits we describe each circuit as a graph and use graph neural networks (GNNs) to learn node embeddings. We show that pretraining GNNs on prediction of output node voltages can encourage learning representations that can be adapted to new unseen topologies or prediction of new circuit level properties.
  arXiv  Detail & Related papers  (2022-03-29T21:18:47Z)
• Data-driven emergence of convolutional structure in neural networks [83.4920717252233]
  We show how fully-connected neural networks solving a discrimination task can learn a convolutional structure directly from their inputs. By carefully designing data models, we show that the emergence of this pattern is triggered by the non-Gaussian, higher-order local structure of the inputs.
  arXiv  Detail & Related papers  (2022-02-01T17:11:13Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.