mlsanityData to 14 Oct 2026

CircuitKIT : Circuit Discovery, Evaluation, and Application Toolkit for Mechanistic Interpretability

arXiv 2607.19317 ↗First posted 21 Jul 2026cs.LG cs.CL cs.ETcited by 0

On the map: Mechanistic interpretability → Circuits and mechanistic analysis

Abstract (arXiv)

Circuit analysis can support not only model explanation but also downstream interventions such as pruning, editing, steering, and selective fine-tuning. However, conducting such analyses currently requires stitching together separate implementations for discovery, evaluation, and intervention, as well as hand-authoring the contrastive prompts required by many discovery methods. This fragmentation makes methods difficult to compare and limits their application beyond canonical tasks. We introduce CircuitKIT, a source-available library that connects the circuit-analysis workflow through a typed, serializable representation. CircuitKIT provides a suite of discovery algorithms, declarative interfaces for mapping structured data into discovery tasks, complementary circuit diagnostics, and downstream application modules. Together, these components provide common infrastructure for conducting and comparing circuit analyses. The library, examples, notebooks, and documentation are released at https://github.com/Lexsi-Labs/CircuitKIT .

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Circuits and mechanistic analysis, one of 13 directions in Mechanistic interpretability. Builds on 27 of its 36 references that are in the map. First posted on arXiv 21 Jul 2026.

Builds on

27 of 36 references.

Closest ancestors · 11

Same direction, newest first

  1. 2026IBCircuit: Towards Holistic Circuit Discovery with Information Bottleneck26 Feb
  2. 2025Circuit Stability Characterizes Language Model Generalization30 May
  3. 2025MIB: A Mechanistic Interpretability Benchmark17 Apr
  4. 2025EAP-GP: Mitigating Saturation Effect in Gradient-based Automated Circuit Identification7 Feb
  5. 2025Position-aware Automatic Circuit Discovery7 Feb
  6. 2024InterpBench: Semi-Synthetic Transformers for Evaluating Mechanistic Interpretability Techniques19 Jul
  7. 2024Transformer Circuit Faithfulness Metrics are not Robust11 Jul
  8. 2024Efficient Automated Circuit Discovery in Transformers using Contextual Decomposition1 Jul
  9. 2023Attribution Patching Outperforms Automated Circuit Discovery16 Oct
  10. 2023Towards Automated Circuit Discovery for Mechanistic Interpretability28 Apr
  11. 2022Interpretability in the Wild: a Circuit for Indirect Object Identification in GPT-2 small1 Nov

Other references · 16

  1. 2017Axiomatic Attribution for Deep Networkscited by 3,384
  2. 2021TruthfulQA: Measuring How Models Mimic Human Falsehoodscited by 2,975
  3. 2022Locating and Editing Factual Associations in GPTcited by 2,691
  4. 2024Refusal in Language Models Is Mediated by a Single Directioncited by 1,032
  5. 2022Mass-Editing Memory in a Transformercited by 1,001
  6. 2020StereoSet: Measuring stereotypical bias in pretrained language modelscited by 924
  7. 2023LLM-Pruner: On the Structural Pruning of Large Language Modelscited by 650
  8. 2020Captum: A unified and generic model interpretability library for PyTorchcited by 547
  9. 2024The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearningcited by 509
  10. 2024Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Modelscited by 435
  11. 2023How does GPT-2 compute greater-than?: Interpreting mathematical abilities in a pre-trained language modelcited by 202
  12. 2024Have Faith in Faithfulness: Going Beyond Circuit Overlap When Finding Model Mechanismscited by 149
4 more
  1. 2026Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language Modelscited by 27
  2. 2025Interpretability as Alignment: Making Internal Understanding a Design Principlecited by 3
  3. 2026Faithfulness to Refusal: A Causal Audit of Neuron Selectorscited by 2
  4. 2026Forgetting That Sticks: Quantization-Permanent Unlearning via Circuit Attributioncited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Many Circuits, One Mechanism: Input Variation and Evaluation Granularity in Circuit Discovery
  2. 2026Rethinking Circuit Evaluation: Do Circuits Explain Model Errors?
  3. 2025Circuit Insights: Towards Interpretability Beyond Activations
  4. 2026Demystifying Variance in Circuit Discovery of LLMs
  5. 2026Formal Mechanistic Interpretability: Automated Circuit Discovery with Provable Guarantees
  6. 2024Automatically Identifying Local and Global Circuits with Linear Computation Graphs
  7. 2024The Computational Complexity of Circuit Discovery for Inner Interpretability
  8. 2026From Circuit Evidence to Mechanistic Theory: An Inductive Logic Approach