Paper Citation Record · LEDGER

TopK Language Models

As of 9 August 2026, this Paper Citation Record lists 49 of 49 outbound references and 0 inbound Pith citation observations for arXiv:2506.21468.

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

pith.paper-citation-record.v1
2506.21468 v1

Coverage vector

measured 49 of 49 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-06T22:31:39.220992Z

measured 49 of 49 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

49 of 49 outbound references displayed

verified exact1
verified fuzzy34
unresolved12
parse uncertain0
malformed identifier1
metadata mismatch1

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 789d64b4-69c1-41fb-80dd-51772945f630 · outbound

This paper cites How can we be so dense? the benefits of using highly sparse representations, 2019.

TopK Language Models How can we be so dense? the benefits of using highly sparse representations, 2019

Reference 1

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:44.922624Z

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 252a90f8-e166-4ccf-8aa8-0427bd7175dd · outbound

This paper cites Generating long sequences with sparse transformers, 2019.

TopK Language Models Generating long sequences with sparse transformers, 2019

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-06T22:31:34.787577Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 674b7941-d134-4ec2-b182-617c4b245cb5 · outbound

This paper cites Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge.

TopK Language Models Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-06T22:31:34.894657Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7a6bcd59-399b-4e81-a450-405ea058ef91 · outbound

This paper cites [Full Post] Progress Update #1 from the GDM Mech Interp Team.

TopK Language Models [Full Post] Progress Update #1 from the GDM Mech Interp Team

Reference 4

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:44.790819Z

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 ba9b874e-ce44-4514-9642-ba67d7764f98 · outbound

This paper cites Sparse autoencoders find highly interpretable features in language models, 2023.

TopK Language Models Sparse autoencoders find highly interpretable features in language models, 2023

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-06T22:31:35.158355Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f0204e56-9a10-4e7c-947b-1d805f936fb0 · outbound

This paper cites Rigging the lottery: Making all tickets winners.

TopK Language Models Rigging the lottery: Making all tickets winners

Reference 6

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:44.685373Z

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 30ea9f6d-00f7-4f49-964f-0f3dc6a16879 · outbound

This paper cites Sparsity in transformers: A systematic literature review.

TopK Language Models Sparsity in transformers: A systematic literature review

Reference 7

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:44.560982Z

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 72d4a652-94cd-4b34-a2aa-d89f288ff9b0 · outbound

This paper cites Detecting hallucinations in large language models using semantic entropy.

TopK Language Models Detecting hallucinations in large language models using semantic entropy

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-06T22:31:35.530869Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6f137638-06ec-406b-9ce0-061a3bf57b53 · outbound

This paper cites Scaling and evaluating sparse autoencoders, 2024.

TopK Language Models Scaling and evaluating sparse autoencoders, 2024

Reference 9

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:44.452327Z

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 f4e0c28a-7209-4240-955b-333c6e0ea7da · outbound

This paper cites The language model evaluation harness, 2024.

TopK Language Models The language model evaluation harness, 2024

Reference 10

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:44.370392Z

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 ee4dd635-db57-410b-9324-1787e68f3a5c · outbound

This paper cites Causal Abstractions of Neural Networks.

TopK Language Models Causal Abstractions of Neural Networks

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:44.255996Z

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 04f5fb92-8a51-49c8-b8d4-d99b82567568 · outbound

This paper cites The State of Sparse Training in Deep Reinforcement Learning.

TopK Language Models The State of Sparse Training in Deep Reinforcement Learning

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:44.158063Z

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 aa35ef23-a1e2-4365-8576-b6daf9a5c923 · outbound

This paper cites Memory-efficient transformers via top-k attention, 2021.

TopK Language Models Memory-efficient transformers via top-k attention, 2021

Reference 13

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:44.047952Z

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 3f7984cd-049b-4bda-86f3-9384fd31b028 · outbound

This paper cites Llama scope: Extracting millions of features from llama-3.1-8b with sparse autoencoders, 2024.

TopK Language Models Llama scope: Extracting millions of features from llama-3.1-8b with sparse autoencoders, 2024

Reference 14

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:43.933604Z

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 cdb4a400-794b-422a-b9db-23e9e5554868 · outbound

This paper cites Hindupur, Ekdeep Singh Lubana, Thomas Fel, and Demba Ba.

TopK Language Models Hindupur, Ekdeep Singh Lubana, Thomas Fel, and Demba Ba

Reference 15

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:43.826679Z

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 93a23817-5ae8-4063-9a61-cc7a57c7407d · outbound

This paper cites Two Sparsities Are Better Than One: Unlocking the Performance Benefits of Sparse-Sparse Networks.

TopK Language Models Two Sparsities Are Better Than One: Unlocking the Performance Benefits of Sparse-Sparse Networks

Reference 16

Resolution: verified exact; local_arxiv, observed 2026-08-06T22:31:39.565510Z

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 431d5c70-f9cd-434d-b265-d0c8d173ceb6 · outbound

This paper cites How llms learn: Tracing internal representations with sparse autoencoders, 2025.

TopK Language Models How llms learn: Tracing internal representations with sparse autoencoders, 2025

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:43.727068Z

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 d3c0bed0-dcef-4bad-a2a3-7229280fc0ad · outbound

This paper cites Sparse is enough in scaling transformers, 2021.

TopK Language Models Sparse is enough in scaling transformers, 2021

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:43.615792Z

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 d7578cb5-abb8-45ff-b7d9-85726b9abc70 · outbound

This paper cites Top-KAST: Top-K Always Sparse Training.

TopK Language Models Top-KAST: Top-K Always Sparse Training

Reference 19

Resolution: metadata mismatch; local_arxiv, observed 2026-08-06T22:31:39.411734Z

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 4f91072b-f53e-4f6f-b9bf-d3d8d7b64d13 · outbound

This paper cites Saebench: A comprehensive benchmark for sparse autoen- coders in language model interpretability, 2025.

TopK Language Models Saebench: A comprehensive benchmark for sparse autoen- coders in language model interpretability, 2025

Reference 20

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:43.505754Z

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 1d8d7c24-9783-465d-9c80-be28083f430a · outbound

This paper cites Concept steerers: Leveraging k-sparse autoencoders for controllable generations, 2025.

TopK Language Models Concept steerers: Leveraging k-sparse autoencoders for controllable generations, 2025

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:43.327447Z

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 88cbb3b3-8ac3-4ba0-8158-3b1251dba858 · outbound

This paper cites Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in Natural Language Generation.

TopK Language Models Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in Natural Language Generation

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:43.166195Z

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 3edfc9c7-94f6-4281-a590-880224887ad8 · outbound

This paper cites Soft Threshold Weight Reparameterization for Learnable Sparsity.

TopK Language Models Soft Threshold Weight Reparameterization for Learnable Sparsity

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:43.002177Z

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 a05d54fd-ae7b-4413-b870-4cecc3413d54 · outbound

This paper cites Sparse autoencoders do not find canonical units of analysis, 2025.

TopK Language Models Sparse autoencoders do not find canonical units of analysis, 2025

Reference 24

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:42.807089Z

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 5f4fba51-bb4a-4ebd-9a82-f17b303fec40 · outbound

This paper cites Gemma scope: Open sparse autoencoders everywhere all at once on gemma 2, 2024.

TopK Language Models Gemma scope: Open sparse autoencoders everywhere all at once on gemma 2, 2024

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-06T22:31:37.056013Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0e04cee2-46f0-46d1-ade5-282892edb508 · outbound

This paper cites Decoupled Weight Decay Regularization.

TopK Language Models Decoupled Weight Decay Regularization

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-06T22:31:37.178870Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dfa20b76-4411-44f4-807f-6be32170786e · outbound

This paper cites Learning sparse neural networks through l_0 regularization.

TopK Language Models Learning sparse neural networks through l_0 regularization

Reference 27

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:42.623753Z

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 8a792f02-b2c6-45e8-b471-bb1f4f042e41 · outbound

This paper cites Fineweb-edu: the finest collection of educational content, 2024.

TopK Language Models Fineweb-edu: the finest collection of educational content, 2024

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-06T22:31:37.336793Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c81c18f1-a5e1-419b-878a-9a4b2a29115b · outbound

This paper cites Winner-take-all autoencoders.

TopK Language Models Winner-take-all autoencoders

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:42.399846Z

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 cf1a13db-df4e-4f85-8d87-3a1f51a45d56 · outbound

This paper cites Pointer Sentinel Mixture Models.

TopK Language Models Pointer Sentinel Mixture Models

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:42.252639Z

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 6cc3d4c0-c29f-45dc-b26e-406dba71cc93 · outbound

This paper cites Can a suit of armor conduct electricity? a new dataset for open book question answering.

TopK Language Models Can a suit of armor conduct electricity? a new dataset for open book question answering

Reference 31

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:42.067358Z

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 69e9c625-3b6b-45ee-883a-cc68e008da50 · outbound

This paper cites Nguyen, Madeleine Gibescu, Antonio Liotta, and et al.

TopK Language Models Nguyen, Madeleine Gibescu, Antonio Liotta, and et al

Reference 32

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:41.886349Z

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 2a719d09-833e-4914-8ee3-0b74cd9203b5 · outbound

This paper cites The LAMBADA dataset: Word prediction requiring a broad discourse context.

TopK Language Models The LAMBADA dataset: Word prediction requiring a broad discourse context

Reference 33

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:41.688817Z

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 6dd0f983-9ab7-4e9c-8faf-5f2e127b6e35 · outbound

This paper cites Sparse autoencoders trained on the same data learn different features, 2025.

TopK Language Models Sparse autoencoders trained on the same data learn different features, 2025

Reference 34

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:41.469546Z

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 0754bfa7-a587-49c0-ab97-1f49c4ce81b1 · outbound

This paper cites Automatically interpreting millions of features in large language models, 2024.

TopK Language Models Automatically interpreting millions of features in large language models, 2024

Reference 35

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:41.268686Z

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 a1125760-f628-4ca2-9ebb-4d6f608633bf · outbound

This paper cites The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scale.

TopK Language Models The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scale

Reference 36

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:41.035447Z

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 d02c0341-8289-4f6a-9e93-7c948279a759 · outbound

This paper cites Improving dictionary learning with gated sparse autoencoders, 2024.

TopK Language Models Improving dictionary learning with gated sparse autoencoders, 2024

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-06T22:31:37.972247Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation acedd645-2f42-434a-8ef1-aebff0af90bb · outbound

This paper cites Winogrande: an adversarial winograd schema challenge at scale.

TopK Language Models Winogrande: an adversarial winograd schema challenge at scale

Reference 38

Resolution: unresolved; no resolver link, observed 2026-08-06T22:31:38.142907Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8f6dab73-abb4-4500-b8d5-f326759064a8 · outbound

This paper cites Taking features out of superposition with sparse autoencoders.

TopK Language Models Taking features out of superposition with sparse autoencoders

Reference 39

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:40.851527Z

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 ed8b5f35-897e-4c8d-9f38-e00bfacbbac6 · outbound

This paper cites Outrageously large neural networks: The sparsely-gated mixture-of-experts layer, 2017.

TopK Language Models Outrageously large neural networks: The sparsely-gated mixture-of-experts layer, 2017

Reference 40

Resolution: unresolved; no resolver link, observed 2026-08-06T22:31:38.370614Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1ddad81e-ea7a-41e9-b420-c9e5d64197bd · outbound

This paper cites A survey on sparse autoencoders: Interpreting the internal mechanisms of large language models, 2025.

TopK Language Models A survey on sparse autoencoders: Interpreting the internal mechanisms of large language models, 2025

Reference 41

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:40.670077Z

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 9c0496bf-71fc-42ea-a76c-b215fc6f7781 · outbound

This paper cites Codebook Features: Sparse and Discrete Interpretability for Neural Networks.

TopK Language Models Codebook Features: Sparse and Discrete Interpretability for Neural Networks

Reference 42

Resolution: unresolved; no resolver link, observed 2026-08-06T22:31:38.680344Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e1b8015f-8dca-48ee-a6d5-94e0197c7317 · outbound

This paper cites Daniel Freeman, Theodore R.

TopK Language Models Daniel Freeman, Theodore R

Reference 43

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:40.462326Z

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 4e902680-9768-47c0-aae7-a1a9b5f98c7c · outbound

This paper cites LLaMA: Open and Efficient Foundation Language Models.

TopK Language Models LLaMA: Open and Efficient Foundation Language Models

Reference 44

Resolution: unresolved; no resolver link, observed 2026-08-06T22:31:38.890079Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3ea79e63-1c97-4e4a-ad6c-80dc996c16ab · outbound

This paper cites Meta Lingua: A minimal PyTorch LLM training library, 2024.

TopK Language Models Meta Lingua: A minimal PyTorch LLM training library, 2024

Reference 45

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:40.282755Z

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 373449a0-f1a3-45e0-b63a-24c6155ab499 · outbound

This paper cites Tracking the feature dynamics in llm training: A mechanistic study, 2025.

TopK Language Models Tracking the feature dynamics in llm training: A mechanistic study, 2025

Reference 46

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:40.093824Z

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 b26d6674-e1c9-49fd-a3d2-63892221c21a · outbound

This paper cites STEP: Staged parameter-efficient pre-training for large language models.

TopK Language Models STEP: Staged parameter-efficient pre-training for large language models

Reference 47

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:39.921354Z

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 4a13895a-5f6b-42e3-a2c2-6056c77dd21a · outbound

This paper cites HellaSwag: Can a machine really finish your sentence? In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics , pages 4791–4800, 2019.

TopK Language Models HellaSwag: Can a machine really finish your sentence? In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics , pages 4791–4800, 2019

Reference 48

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T22:31:39.727732Z

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 757b2521-884d-4004-a072-b40518b8c712 · outbound

This paper cites PyTorch FSDP: Experiences on Scaling Fully Sharded Data Parallel.

TopK Language Models PyTorch FSDP: Experiences on Scaling Fully Sharded Data Parallel

Reference 49

Resolution: malformed identifier; no resolver link, observed 2026-08-06T22:31:39.220992Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

No inbound Pith citation observations are available.