The paper introduces Concept-Targeted Attribution (CTA), a method that trains attribution graphs to explain the emergence of internal concept representations in language models, rather than just the final token prediction. CTA produces probe-specific circuits that reveal which internal computations drive a linear probe’s accuracy, and cross-layer transcoders demonstrate that these graphs contain predictive structure across multiple concept categories. Causal ablations show that probe-targeted and logit-targeted graphs capture distinct mechanisms, with probe-relevant features affecting internal concept scores and logit-relevant features altering generated tokens.
By Vedant Palit, Florent Draye, Terry Jingchen Zhang, Bernhard Sch\"olkopf, Zhijing Jin
arXiv:2608. 02632v1 Announce Type: new Abstract: Circuit tracing is an exciting technique for revealing the internal computation of language models, but it requires a time-intensive manual step of grouping individual features or MLP neurons into supernodes.
By Ameen Patel, Max Zhang, Nathan Hu
arXiv:2607. 18921v1 Announce Type: cross Abstract: Circuit extraction identifies a small set of model components whose presence preserves a target behavior under ablation, and the resulting circuit is often read as the mechanism behind that behavior.
By Yang Sheng, Jie Fu
arXiv:2603. 21014v2 Announce Type: replace Abstract: Mechanistic interpretability seeks to understand how Large Language Models (LLMs) represent and process information.
By Florent Draye, Vedant Palit, Abir Harrasse, Tung-Yu Wu, Jiarui Liu, Punya Syon Pandey, Roderick Wu, Chih-Hao Hsu, Terry Jingchen Zhang, Zhijing Jin, Bernhard Sch\"olkopf
arXiv:2608. 08536v1 Announce Type: new Abstract: Circuit localization is a mechanistic interpretability task whose goal is to identify a sparse subgraph of a transformer's computation graph sufficient to reproduce a particular behavior.
By Chester Tan, Moritz Lampert, Courtney Maynard, Ankit Ramakrishnan, Tina Eliassi-Rad, Ingo Scholtes
arXiv:2608. 09117v1 Announce Type: new Abstract: Probabilistic Circuits (PCs) are tractable generative models whose internal nodes encode a hierarchy of probabilistic sum- maries over different variable scopes.
By Bhumika K, Vidhya S, Narayanan C Krishnan
ReDIL-GNN is a framework for resynthesis domain‑incremental learning in circuit graph neural networks. It adapts a fixed prediction or representation head as new synthesis styles appear and evaluates retention across all previously seen domains. The method introduces the Resynthesis Adaptability Index (RAI), a pre‑adaptation score that combines adaptation need, source‑equivalence recoverability, structural coverage, and update compatibility to decide whether to adapt, reuse, or defer updates.
By Rupesh Raj Karn, Johann Knechtel, Ozgur Sinanoglu
arXiv:2606. 08236v1 Announce Type: cross Abstract: As large language models are increasingly deployed in high-stakes settings, there is a growing need for tools that audit not only model outputs but also the internal computations that produce them.
By Hyunjin Cho, Youngji Roh, Jaehyung Kim
Diff Mining is a framework that identifies what a finetuned language model has learned by comparing its logits to those of its base model. It extracts per-context logit differences on a reference corpus and aggregates them into an interpretable token set using either a Top‑K frequency method or Non‑negative Matrix Factorization. The approach outperforms existing model‑diffing methods in domain detection and bias identification, and it requires only access to output logits, making it scalable to large models.
By Greg Kocher, Robert West, Cl\'ement Dumas, Julian Minder
arXiv:2607. 27290v1 Announce Type: new Abstract: Modern telecommunication, cloud, and microservice systems emit correlated alarm cascades when components fail.
By Lei Zan, Keli Zhang, Shifeng Xie, Jiale Zheng, Zehao Xiao, Zhiwei Dong, Ke Zhang, Ruichu Cai, Malik Tiomoko, Lujia Pan
arXiv:2606. 09899v1 Announce Type: cross Abstract: A central goal of mechanistic interpretability is to identify which internal components causally drive a language model's behavior.
By Luyang Zhang, Jialu Wang
arXiv:2607. 19317v1 Announce Type: new Abstract: Circuit analysis can support not only model explanation but also downstream interventions such as pruning, editing, steering, and selective fine-tuning.
By Pratinav Seth, Hem Gosalia, Aditya Kasliwal, Vinay Kumar Sankarapu