From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers | Paper Citation Record

Observation 64ab2eca-44f6-48a9-99fd-20e5b87d9c7c · outbound

This paper cites Interpretable machine learning–a brief history, state-of-the-art and challenges.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Interpretable machine learning–a brief history, state-of-the-art and challenges

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.146049Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b643256d-d1e9-4627-926b-87ffc578a1fd · outbound

This paper cites Explainable ai: A review of machine learning interpretability methods.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Explainable ai: A review of machine learning interpretability methods

Reference 2

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.793204Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation ff3d312b-6801-4a5e-bd20-8d9fd9fc6294 · outbound

This paper cites Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.157002Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5e2b43f0-461b-4252-a661-6d930a520d88 · outbound

This paper cites Understanding Neural Networks Through Deep Visualization.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Understanding Neural Networks Through Deep Visualization

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.162267Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 642212fa-24cd-4ad3-b00d-3bff67f46833 · outbound

This paper cites Visualizing deep neural network decisions: Prediction difference analysis.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Visualizing deep neural network decisions: Prediction difference analysis

Reference 5

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.776931Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 4b9ac762-14d7-4e77-8bac-2a686f14ad36 · outbound

This paper cites Grad-cam: Visual explanations from deep networks via gradient-based localization.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Grad-cam: Visual explanations from deep networks via gradient-based localization

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.172216Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 20b3044b-6645-4246-9b13-bdd5ea2172f9 · outbound

This paper cites Axiomatic attribution for deep networks.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Axiomatic attribution for deep networks

Reference 7

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.751452Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 237e3feb-ef2f-46de-8d7d-2f62a0885404 · outbound

This paper cites Attention is all you need.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Attention is all you need

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.736195Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 9761cfd0-fe32-434d-a5ea-4fd9d92ba08f · outbound

This paper cites Rethinking Interpretability in the Era of Large Language Models.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Rethinking Interpretability in the Era of Large Language Models

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.186664Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e7b4bd07-1b98-45f0-ab26-e81f09eee511 · outbound

This paper cites Transformer feed-forward layers are key-value memories.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Transformer feed-forward layers are key-value memories

Reference 10

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.720636Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 0c796c4d-e5e5-4d7e-a76c-5e6daa386b1a · outbound

This paper cites Transformer feed-forward layers build predictions by promoting concepts in the vocabulary space.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Transformer feed-forward layers build predictions by promoting concepts in the vocabulary space

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.703794Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 5a469460-6d1a-4e4e-b51d-f9904a857e38 · outbound

This paper cites The emergence of number and syntax units in lstm language models.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers The emergence of number and syntax units in lstm language models

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.687969Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 221d03ec-4c1d-4460-8928-c38e345c0c20 · outbound

This paper cites Locating and editing factual associations in gpt.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Locating and editing factual associations in gpt

Reference 13

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.671558Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 1d3f0020-5787-4ebc-a371-fd24cf007610 · outbound

This paper cites Mass-Editing Memory in a Transformer.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Mass-Editing Memory in a Transformer

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.211980Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3fe043fa-b81c-43bd-bba2-69f7f47568d6 · outbound

This paper cites Lima: Less is more for alignment.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Lima: Less is more for alignment

Reference 15

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.655741Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 2691784c-8244-46e8-9204-be5d940657d0 · outbound

This paper cites HarmBench: A standardized evaluation framework for automated red teaming and robust refusal.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers HarmBench: A standardized evaluation framework for automated red teaming and robust refusal

Reference 16

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.639436Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 819be478-7c94-44f6-b8e0-d3bf80a4262a · outbound

This paper cites Training Verifiers to Solve Math Word Problems.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Training Verifiers to Solve Math Word Problems

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.226502Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a8c62cfb-af2f-4037-8945-8fe333d55559 · outbound

This paper cites A mathematical framework for transformer circuits.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers A mathematical framework for transformer circuits

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.232206Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f6cce585-25e4-44ca-ae2b-ff260f21d0e8 · outbound

This paper cites Efficient Estimation of Word Representations in Vector Space.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Efficient Estimation of Word Representations in Vector Space

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.241939Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cd49423f-e7be-4217-8a54-32e790c4e333 · outbound

This paper cites Daniel Freeman, Theodore R.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Daniel Freeman, Theodore R

Reference 20

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.597064Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 83b3e45c-ae52-4120-a566-991f529ade7a · outbound

This paper cites Gemma Scope: Open Sparse Autoencoders Everywhere All At Once on Gemma 2.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Gemma Scope: Open Sparse Autoencoders Everywhere All At Once on Gemma 2

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.251181Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c9626cbd-ac7f-4a70-b9e8-ea500237e2fb · outbound

This paper cites Linear Representations of Sentiment in Large Language Models.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Linear Representations of Sentiment in Large Language Models

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.255950Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5ac3ff6e-50cb-46cf-a1a5-3aba4bea503c · outbound

This paper cites The Geometry of Truth: Emergent Linear Structure in Large Language Model Representations of True/False Datasets.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers The Geometry of Truth: Emergent Linear Structure in Large Language Model Representations of True/False Datasets

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.260503Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3fbd40d7-8645-4ec1-95d8-3521f64a2617 · outbound

This paper cites Steering Llama 2 via Contrastive Activation Addition.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Steering Llama 2 via Contrastive Activation Addition

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.265177Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b2124474-298c-46db-94ea-6adb117f5cf8 · outbound

This paper cites Improving Activation Steering in Language Models with Mean-Centring.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Improving Activation Steering in Language Models with Mean-Centring

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.270009Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d569c0c0-f6bf-4ada-aa56-97d22086cf4e · outbound

This paper cites Refusal in Language Models Is Mediated by a Single Direction.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Refusal in Language Models Is Mediated by a Single Direction

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.275006Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a58f1d1e-39a8-4eb5-bd90-16e91e99310b · outbound

This paper cites interpreting gpt: the logit lens.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers interpreting gpt: the logit lens

Reference 27

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.580827Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 6919faa0-7e13-41b7-8ac2-92568ab038c5 · outbound

This paper cites Investigating gender bias in language models using causal mediation analysis.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Investigating gender bias in language models using causal mediation analysis

Reference 28

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.565264Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation bcdc1773-2f24-431e-b9e6-994e4b714c29 · outbound

This paper cites Localizing Model Behavior with Path Patching.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Localizing Model Behavior with Path Patching

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.289390Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3e7f17ba-d12a-4f79-8e33-91b390a63f50 · outbound

This paper cites Inducing causal structure for interpretable neural networks.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Inducing causal structure for interpretable neural networks

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.549853Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 909a3951-0689-420c-8119-5d57319de618 · outbound

This paper cites Transformerlens.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Transformerlens

Reference 31

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.534510Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation db1964c7-2d63-417f-85b7-abbfe89e212b · outbound

This paper cites Vit prisma: A mechanistic interpretability library for vision transformers.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Vit prisma: A mechanistic interpretability library for vision transformers

Reference 32

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.519580Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation d3fe72d7-9fc3-49cc-a5bd-fa86c1c47c03 · outbound

This paper cites How do Large Language Models Handle Multilingualism?.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers How do Large Language Models Handle Multilingualism?

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.308331Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 99bd0b90-0298-4efb-97cc-60a315ee15be · outbound

This paper cites Do Llamas Work in English? On the Latent Language of Multilingual Transformers.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Do Llamas Work in English? On the Latent Language of Multilingual Transformers

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.313609Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fcb8239f-9043-4c1c-b6d0-7581405ae700 · outbound

This paper cites Llama Scope: Extracting Millions of Features from Llama-3.1-8B with Sparse Autoencoders.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Llama Scope: Extracting Millions of Features from Llama-3.1-8B with Sparse Autoencoders

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.319117Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0d257a39-dcee-4505-9392-c535afc26dcf · outbound

This paper cites Scaling and evaluating sparse autoencoders.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Scaling and evaluating sparse autoencoders

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.324277Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4d5e9480-8f7b-4658-933d-fb99fa067da3 · outbound

This paper cites Gemma: Open Models Based on Gemini Research and Technology.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Gemma: Open Models Based on Gemini Research and Technology

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.329084Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bb2f4150-01fe-4982-b398-40de9ffdb133 · outbound

This paper cites Towards monosemanticity: Decomposing language models with dictionary learning.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Towards monosemanticity: Decomposing language models with dictionary learning

Reference 38

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.503642Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation b299e770-df4c-4212-aed7-614042f39d1b · outbound

This paper cites Chain-of-thought prompting elicits reasoning in large language models.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Chain-of-thought prompting elicits reasoning in large language models

Reference 39

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.487516Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 73ebeb61-65e3-4b43-a40b-2d146a86ddb3 · outbound

This paper cites A comprehensive survey on test-time adaptation under distribution shifts.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers A comprehensive survey on test-time adaptation under distribution shifts

Reference 40

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.471484Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 3f473369-48c2-4778-8403-99a9eac5c26f · outbound

This paper cites CommonsenseQA: A question answering challenge targeting commonsense knowledge.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers CommonsenseQA: A question answering challenge targeting commonsense knowledge

Reference 41

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.455413Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 687ac887-d8c1-4734-9783-fabfd09df58a · outbound

This paper cites Evaluating Large Language Models Trained on Code.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Evaluating Large Language Models Trained on Code

Reference 42

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.353257Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 17e0b8fc-f7a5-488e-9a38-a31038f39cdc · outbound

This paper cites Program Synthesis with Large Language Models.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Program Synthesis with Large Language Models

Reference 43

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.357976Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d7a1ab93-b49f-47ca-ba9d-fa00e0a9023e · outbound

This paper cites Is your code generated by chatGPT really correct? rigorous evaluation of large language models for code generation.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Is your code generated by chatGPT really correct? rigorous evaluation of large language models for code generation

Reference 44

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.439661Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 8075964b-0708-4cb9-b11e-d80c6d4c5c93 · outbound

This paper cites Judging llm-as-a-judge with mt-bench and chatbot arena.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Judging llm-as-a-judge with mt-bench and chatbot arena

Reference 45

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.423667Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 4b7d8c2e-f06d-4c25-bc2d-ddc1ad5348cf · outbound

This paper cites A framework for few-shot language model evaluation, 07 2024.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers A framework for few-shot language model evaluation, 07 2024

Reference 46

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.409202Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 32333543-dfa8-417c-8d82-0f8846f9c097 · outbound

This paper cites Interpretability dreams.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Interpretability dreams

Reference 47

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.394726Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 9064121e-3617-45a1-adf7-3cd50690821e · outbound

This paper cites Llama 2: Open Foundation and Fine-Tuned Chat Models.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Llama 2: Open Foundation and Fine-Tuned Chat Models

Reference 48

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.381437Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ce5b0fc8-b6da-4bbe-b3ed-d1147dba9c51 · outbound

This paper cites Qwen Technical Report.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Qwen Technical Report

Reference 49

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.385930Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f2c96340-ff62-45bd-8b02-11fe58fe6bf3 · outbound

This paper cites Universal and Transferable Adversarial Attacks on Aligned Language Models.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Universal and Transferable Adversarial Attacks on Aligned Language Models

Reference 50

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.390705Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8a798688-a531-43f5-8999-672f930d95ff · outbound

This paper cites An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale

Reference 51

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.395395Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5f00f00d-7a0a-4be8-81eb-a06372aa70ed · outbound

This paper cites Vision transformers need registers.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Vision transformers need registers

Reference 52

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.380338Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation cefe7a3d-2aab-41ed-bbbe-6e8af1ef90a2 · outbound

This paper cites Visualizing and understanding convolutional networks.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Visualizing and understanding convolutional networks

Reference 53

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.365989Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 2137c90c-b7f3-4931-998d-6da6021f36ea · outbound

This paper cites Network dissec- tion: Quantifying interpretability of deep visual representations.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Network dissec- tion: Quantifying interpretability of deep visual representations

Reference 54

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.351476Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation d88e2d76-bb37-4b71-a6f0-b9b7248221be · outbound

This paper cites Interpreting deep visual representa- tions via network dissection.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Interpreting deep visual representa- tions via network dissection

Reference 55

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.335811Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation aa756b2a-7760-441c-a14a-5d5c9283c4ee · outbound

This paper cites Interpretable basis decomposition for visual explanation.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Interpretable basis decomposition for visual explanation

Reference 56

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.320829Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 4fa9617f-09e8-45bb-aa96-e094ef2ced7e · outbound

This paper cites Understanding the role of individual units in a deep neural network.Proceedings of the National Academy of Sciences, 117(48):30071–30078, 2020.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Understanding the role of individual units in a deep neural network.Proceedings of the National Academy of Sciences, 117(48):30071–30078, 2020

Reference 57

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.305156Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation e2db74ee-c989-4540-b24d-8c16022b83bf · outbound

This paper cites Dissecting recall of factual associations in auto-regressive language models.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Dissecting recall of factual associations in auto-regressive language models

Reference 58

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.290059Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation a0d43275-b1e8-40e1-b4b5-f1e6e0a668fe · outbound

This paper cites Does localization inform editing? surprising differences in causality-based localization vs.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Does localization inform editing? surprising differences in causality-based localization vs

Reference 59

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.274066Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation a2e6a373-50b8-476c-8736-978ca4eba91b · outbound

This paper cites Editing common sense in transformers.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Editing common sense in transformers

Reference 60

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.258520Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 3e7da4eb-9637-4224-b636-9d2aac054a2d · outbound

This paper cites Massive editing for large language models via meta learning.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Massive editing for large language models via meta learning

Reference 61

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.242990Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation b549e9ad-8aed-47db-874e-adf4fadb6513 · outbound

This paper cites Pmet: Precise model editing in a transformer.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Pmet: Precise model editing in a transformer

Reference 62

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.227618Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation f29ad2ba-d8dd-4f5e-ad0b-33c035e61128 · outbound

This paper cites Journey to the center of the knowledge neurons: Discoveries of language-independent knowledge neurons and degenerate knowledge neurons.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Journey to the center of the knowledge neurons: Discoveries of language-independent knowledge neurons and degenerate knowledge neurons

Reference 63

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.213111Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation d1b6891a-83c8-4bc3-b3b1-41f22a11dcb6 · outbound

This paper cites Are Neural Nets Modular? Inspecting Functional Modularity Through Differentiable Weight Masks.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Are Neural Nets Modular? Inspecting Functional Modularity Through Differentiable Weight Masks

Reference 64

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.457340Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e16d825b-4161-4458-b8d0-0e7b859f6f3b · outbound

This paper cites Sparse Interventions in Language Models with Differentiable Masking.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Sparse Interventions in Language Models with Differentiable Masking

Reference 65

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.462472Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 28b192fa-5f25-4c66-a537-0bd45eb0313a · outbound

This paper cites Knowledge Neurons in Pretrained Transformers.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Knowledge Neurons in Pretrained Transformers

Reference 66

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.467637Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bc6a5fd1-ef38-48c3-a8c3-a80f97a42aef · outbound

This paper cites Finding and editing multi- modal neurons in pre-trained transformers.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Finding and editing multi- modal neurons in pre-trained transformers

Reference 67

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.197087Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation f6a365e1-e286-470a-9e25-1a23d00a9f89 · outbound

This paper cites Towards neuron attributions in multi-modal large language models.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Towards neuron attributions in multi-modal large language models

Reference 68

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.180824Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 94bf6f2b-6176-4f0c-922b-f484219d0d79 · outbound

This paper cites Retrieval Head Mechanistically Explains Long-Context Factuality.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Retrieval Head Mechanistically Explains Long-Context Factuality

Reference 69

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.481918Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9aed7111-a7da-4ae8-ba57-ed04aa60d13c · outbound

This paper cites Successor Heads: Recurring, Interpretable Attention Heads In The Wild.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Successor Heads: Recurring, Interpretable Attention Heads In The Wild

Reference 70

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.486631Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b586e7ea-7c1e-4318-93cc-4662c38e4816 · outbound

This paper cites Sparse Autoencoders Find Highly Interpretable Features in Language Models.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Sparse Autoencoders Find Highly Interpretable Features in Language Models

Reference 71

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.491909Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7300ee97-b5f2-4e4a-8dd3-616d91825639 · outbound

This paper cites Zoom in: An introduction to circuits.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Zoom in: An introduction to circuits

Reference 72

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.497381Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6d80426b-0b0d-41e9-8963-a3827e1d2793 · outbound

This paper cites Mechanistic Interpretability for AI Safety -- A Review.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Mechanistic Interpretability for AI Safety -- A Review

Reference 73

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.501945Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b99e52d2-27e3-4523-b6f1-919dd739f733 · outbound

This paper cites Open Problems in Mechanistic Interpretability.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Open Problems in Mechanistic Interpretability

Reference 74

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.506840Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6be5256e-4ce6-475a-88c2-b11c428be253 · outbound

This paper cites In-context learning and induction heads.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers In-context learning and induction heads

Reference 75

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.164812Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 8c13d93a-1e39-41d0-a3ea-ae34912b38ab · outbound

This paper cites Interpretability in the wild: a circuit for indirect object identification in GPT-2 small.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Interpretability in the wild: a circuit for indirect object identification in GPT-2 small

Reference 76

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.149290Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation e5519bd1-d3fd-4315-a5cb-0501924c2b6d · outbound

This paper cites Progress measures for grokking via mechanistic interpretability.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Progress measures for grokking via mechanistic interpretability

Reference 77

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.132461Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation ca93db5b-6680-463e-9a74-f5884919918a · outbound

This paper cites Iteration Head: A Mechanistic Study of Chain-of-Thought.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Iteration Head: A Mechanistic Study of Chain-of-Thought

Reference 78

Resolution: unresolved; no resolver link, observed 2026-08-15T19:17:52.525737Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d5636ce6-db96-401c-8c66-602474f4cdd5 · outbound

This paper cites the Golden Gate Bridge.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers the Golden Gate Bridge

Reference 79

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.112770Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 702bfc56-2627-4152-8ce0-b9ec394778f0 · outbound

This paper cites Tabby cat.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Tabby cat

Reference 81

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T19:17:53.096049Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 8c24b08e-4f0d-4a80-89b8-f589c85e364d · outbound

This paper cites an unresolved cited work.

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers Unresolved cited work

Reference 2021

Resolution: unresolved; raw_fallback, observed 2026-08-15T19:17:53.612735Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Coverage vector

Reference resolution

Outbound references

Pith citing papers