Paper Citation Record · LEDGER

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering

As of 8 August 2026, this Paper Citation Record lists 45 of 45 outbound references and 0 inbound Pith citation observations for arXiv:2506.08359.

A citation records a reference. It does not transfer a finding from one paper to another.

pith.paper-citation-record.v1
2506.08359 v3

Coverage vector

measured 45 of 45 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-07T05:19:30.202178Z

measured 45 of 45 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-08T06:32:00.761636+00:00

measured 0 of 0 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links

measured 0 of 1 external citation measurements

A source-named dated measurement, never combined with another source.

Source: cited_works

Reference resolution

45 of 45 outbound references displayed

verified exact0
verified fuzzy26
unresolved19
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 682dde86-4774-459c-8dae-7418cdaaff18 · outbound

This paper cites Viégas, Hanspeter Pfister, and Martin Wattenberg.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Viégas, Hanspeter Pfister, and Martin Wattenberg

Reference 1

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T05:19:30.641408Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 6d7153da-7d78-4441-adcf-7a07e2b50523 · outbound

This paper cites Steering gpt-2-xl by adding an activation vector.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Steering gpt-2-xl by adding an activation vector

Reference 2

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T05:19:30.632036Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 9cd0c6e5-70fd-4754-a2ed-457d4356558e · outbound

This paper cites Steering llama 2 via contrastive activation addition.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Steering llama 2 via contrastive activation addition

Reference 3

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T05:19:30.621830Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation ed5824e2-6c5d-4a93-87b9-e9df474926b4 · outbound

This paper cites Direct preference optimization: Your language model is secretly a reward model.Advances in Neural Information Processing Systems, 36:53728–53741, 2023.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Direct preference optimization: Your language model is secretly a reward model.Advances in Neural Information Processing Systems, 36:53728–53741, 2023

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-07T05:19:30.054145Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 45b52d2d-15c8-471a-b2de-8ce9a1f802ea · outbound

This paper cites Locating and editing factual associations in gpt.Advances in neural information processing systems, 35:17359–17372, 2022.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Locating and editing factual associations in gpt.Advances in neural information processing systems, 35:17359–17372, 2022

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-07T05:19:30.057896Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7eb8e8d5-2c91-46c9-9fc7-679ea25f9da8 · outbound

This paper cites Programming Refusal with Conditional Activation Steering.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Programming Refusal with Conditional Activation Steering

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-07T05:19:30.061438Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cba53e92-636d-477b-a5d3-569ca5ec26c3 · outbound

This paper cites TruthX: Alleviating hallucinations by editing large language models in truthful space.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering TruthX: Alleviating hallucinations by editing large language models in truthful space

Reference 7

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T05:19:30.601859Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 0526728d-d7ea-4bc2-84b4-d3476aacf0b6 · outbound

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

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Refusal in Language Models Is Mediated by a Single Direction

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-07T05:19:30.068919Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 05f4ba7b-2112-4b02-ac9f-af9092261ee9 · outbound

This paper cites Truthflow: Truthful LLM generation via representation flow correction.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Truthflow: Truthful LLM generation via representation flow correction

Reference 9

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T05:19:30.592290Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 3237afe8-e269-4558-9617-b3417eee2b73 · outbound

This paper cites an unresolved cited work.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Unresolved cited work

Reference 10

Resolution: unresolved; raw_fallback, observed 2026-08-07T05:19:30.582773Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 6120db2a-46b1-4c39-856c-b53628a09f36 · outbound

This paper cites Steering knowledge selection behaviours in LLMs via SAE-based representation engineering.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Steering knowledge selection behaviours in LLMs via SAE-based representation engineering

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T05:19:30.573045Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 13d1151f-dd77-47e0-b3f2-ffff6f123dca · outbound

This paper cites Lofit: Localized fine-tuning on LLM representations.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Lofit: Localized fine-tuning on LLM representations

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T05:19:30.564081Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 6d01e4b4-72b9-435b-995e-69b72b3b691f · outbound

This paper cites Induction heads as an essential mechanism for pattern matching in in-context learning.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Induction heads as an essential mechanism for pattern matching in in-context learning

Reference 13

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T05:19:30.554196Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation ff1fb717-3da4-45ec-9b50-2b6dd27db6f9 · outbound

This paper cites In-context Learning and Induction Heads.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering In-context Learning and Induction Heads

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-07T05:19:30.090177Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1928aaf2-3c6e-47ef-b15e-9150072269f2 · outbound

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

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Retrieval Head Mechanistically Explains Long-Context Factuality

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-07T05:19:30.093604Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 084a692f-43d5-4ba7-bf6c-d75f655a7825 · outbound

This paper cites A glitch in the matrix? locating and detecting language model grounding with fakepedia.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering A glitch in the matrix? locating and detecting language model grounding with fakepedia

Reference 16

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T05:19:30.543334Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation a7746e88-e267-4a16-ab21-a2c6f35cccb0 · outbound

This paper cites A Primer on the Inner Workings of Transformer-based Language Models.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering A Primer on the Inner Workings of Transformer-based Language Models

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-07T05:19:30.100608Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b648afde-bc63-4a18-a04f-c5e3d744c08e · outbound

This paper cites Truthfulqa: Measuring how models mimic human falsehoods.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Truthfulqa: Measuring how models mimic human falsehoods

Reference 19

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T05:19:30.533886Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 0abfeed8-dfdd-465f-a77a-1a7a6c0ea2b7 · outbound

This paper cites MQuAKE: Assessing knowledge editing in language models via multi-hop questions.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering MQuAKE: Assessing knowledge editing in language models via multi-hop questions

Reference 20

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T05:19:30.523663Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 6c06eb5b-596e-400d-a1a8-e4daf14066b7 · outbound

This paper cites Hamilton.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Hamilton

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T05:19:30.513454Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 31f0770d-b65e-4a5d-8f75-1c84b5e6e8ed · outbound

This paper cites Entity-based knowledge conflicts in question answering.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Entity-based knowledge conflicts in question answering

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T05:19:30.503154Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation a92fb971-1b28-41cb-9d9a-f24a910c4efb · outbound

This paper cites Why so gullible? enhancing the robustness of retrieval-augmented models against counterfactual noise.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Why so gullible? enhancing the robustness of retrieval-augmented models against counterfactual noise

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T05:19:30.488851Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation f1a68ef2-c266-4c78-bef4-a61e40d18845 · outbound

This paper cites Discovering Language Model Behaviors with Model-Written Evaluations.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Discovering Language Model Behaviors with Model-Written Evaluations

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-07T05:19:30.128361Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1aaa82a5-8eba-468f-a660-25793b5c8297 · outbound

This paper cites Contrastive decoding: Open-ended text generation as optimization.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Contrastive decoding: Open-ended text generation as optimization

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-07T05:19:30.131991Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 71c95c59-7702-4b58-b4d8-9628968457c2 · outbound

This paper cites Glass, and Pengcheng He.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Glass, and Pengcheng He

Reference 26

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T05:19:30.472909Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 4a1de50c-71c6-4266-b82d-a2c93b770081 · outbound

This paper cites In- context sharpness as alerts: An inner representation perspective for hallucination mitigation.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering In- context sharpness as alerts: An inner representation perspective for hallucination mitigation

Reference 27

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T05:19:30.462691Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 134168d1-0bf8-4bdd-9b67-257b4d762f08 · outbound

This paper cites In-context learning creates task vectors.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering In-context learning creates task vectors

Reference 28

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T05:19:30.452959Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 8211caa3-5d71-4ef9-93d1-a9d49dec6726 · outbound

This paper cites Steering Language Models With Activation Engineering.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Steering Language Models With Activation Engineering

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-07T05:19:30.145386Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d72fe0a3-2506-435a-b70c-91bbfadbd3ce · outbound

This paper cites Spectral Editing of Activations for Large Language Model Alignment.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Spectral Editing of Activations for Large Language Model Alignment

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-07T05:19:30.148340Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2a57ad75-c2ad-4fbc-a97a-d6b6733aff18 · outbound

This paper cites Trusting your evidence: Hallucinate less with context-aware decoding.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Trusting your evidence: Hallucinate less with context-aware decoding

Reference 31

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T05:19:30.442153Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 6a2a0585-b7ee-42d5-9638-b634c8a7e74f · outbound

This paper cites Language Models are Few-Shot Learners.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Language Models are Few-Shot Learners

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-07T05:19:30.155432Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 127cda6a-d2c4-4cff-86f1-e70a39549cb9 · outbound

This paper cites GQA: training generalized multi-query transformer models from multi-head checkpoints.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering GQA: training generalized multi-query transformer models from multi-head checkpoints

Reference 33

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T05:19:30.432771Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 53720205-313c-49ba-bb67-dfd2fef56f4e · outbound

This paper cites ConflictBank: A Benchmark for Evaluating the Influence of Knowledge Conflicts in LLM.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering ConflictBank: A Benchmark for Evaluating the Influence of Knowledge Conflicts in LLM

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-07T05:19:30.165086Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 597b7251-8dc8-4e8b-a28f-977889e3539f · outbound

This paper cites Adaptive chameleon or stubborn sloth: Revealing the behavior of large language models in knowledge conflicts.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Adaptive chameleon or stubborn sloth: Revealing the behavior of large language models in knowledge conflicts

Reference 35

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T05:19:30.417093Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 1d86a6e5-824b-40d0-bb41-d85bf2ce674e · outbound

This paper cites Combating misinformation in the age of llms: Opportunities and challenges.AI Magazine, 2023.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Combating misinformation in the age of llms: Opportunities and challenges.AI Magazine, 2023

Reference 36

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T05:19:30.407836Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 229db62a-32f6-4090-99d9-179127be7c88 · outbound

This paper cites When not to trust language models: Investigating effectiveness of parametric and non-parametric memories.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering When not to trust language models: Investigating effectiveness of parametric and non-parametric memories

Reference 37

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T05:19:30.396450Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 53dc8559-cf7b-4154-9f91-b3e70cfab6d4 · outbound

This paper cites Pytorch: An imperative style, high-performance deep learning library.Advances in neural information processing systems, 32, 2019.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Pytorch: An imperative style, high-performance deep learning library.Advances in neural information processing systems, 32, 2019

Reference 38

Resolution: unresolved; no resolver link, observed 2026-08-07T05:19:30.180154Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2fda548e-59aa-4c2c-bfda-0f59f7b8db89 · outbound

This paper cites HuggingFace's Transformers: State-of-the-art Natural Language Processing.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering HuggingFace's Transformers: State-of-the-art Natural Language Processing

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-07T05:19:30.184036Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 12733733-98e5-4040-b41c-88490cb3007b · outbound

This paper cites Evaluating the ripple effects of knowledge editing in language models.Trans.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Evaluating the ripple effects of knowledge editing in language models.Trans

Reference 40

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T05:19:30.379182Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 0799f130-07aa-4ae3-a0b6-9974aeee7b32 · outbound

This paper cites Advanced AI Risk.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Advanced AI Risk

Reference 41

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T05:19:30.368957Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation bc330a0d-1e46-4e88-9db5-514e971cad13 · outbound

This paper cites an unresolved cited work.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Unresolved cited work

Reference 44

Resolution: unresolved; raw_fallback, observed 2026-08-07T05:19:30.358220Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 41100c45-e8f1-4482-aed8-33f29049a1ba · outbound

This paper cites Typically, τ is computed as the (100−p) -th percentile over the collection of all head scores, we useτ= 5for all experiments: τ= Percentile {sl,h}l,h,100−p.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Typically, τ is computed as the (100−p) -th percentile over the collection of all head scores, we useτ= 5for all experiments: τ= Percentile {sl,h}l,h,100−p

Reference 45

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T05:19:30.348885Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 9f8a16d0-61a2-461c-8a4b-e34e0c2ddf9d · outbound

This paper cites linear” and “SqExp.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering linear” and “SqExp

Reference 46

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T05:19:30.337176Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 2b66d5ea-41a7-46b7-9b4b-64f46521bc7c · outbound

This paper cites an unresolved cited work.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Unresolved cited work

Reference 2021

Resolution: unresolved; no resolver link, observed 2026-08-07T05:19:30.120899Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 30b7d024-9f38-4efa-a940-c7279563ef91 · outbound

This paper cites an unresolved cited work.

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering Unresolved cited work

Reference 4901

Resolution: unresolved; no resolver link, observed 2026-08-07T05:19:30.161792Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

No inbound Pith citation observations are available.