Paper Citation Record · LEDGER

Self-Ablating Transformers: More Interpretability, Less Sparsity

As of 18 August 2026, this Paper Citation Record lists 41 of 41 outbound references and 0 inbound Pith citation observations for arXiv:2505.00509.

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

pith.paper-citation-record.v1
2505.00509 v1

Coverage vector

measured 41 of 41 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-16T04:43:45.651719Z

measured 41 of 41 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-18T06:34:40.430872+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

41 of 41 outbound references displayed

verified exact1
verified fuzzy9
unresolved31
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 4ea825f7-0835-4a2d-94b6-fbab139c7034 · outbound

This paper cites write newline.

Self-Ablating Transformers: More Interpretability, Less Sparsity write newline

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.506675Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6af40797-922c-432f-b2ea-332e20faa2b1 · outbound

This paper cites Language Models Can Explain Neurons in Language Models , May 2023.

Self-Ablating Transformers: More Interpretability, Less Sparsity Language Models Can Explain Neurons in Language Models , May 2023

Reference 2

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:43:46.093263Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation a997d328-5dde-4958-bfa8-8085284a4226 · outbound

This paper cites GPT-Neo: Large Scale Autoregressive Language Modeling with Mesh-Tensorflow , March 2021.

Self-Ablating Transformers: More Interpretability, Less Sparsity GPT-Neo: Large Scale Autoregressive Language Modeling with Mesh-Tensorflow , March 2021

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.516143Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 03bbbbab-a3c2-401e-b729-5c48f1369363 · outbound

This paper cites Gradient Routing: Masking Gradients to Localize Computation in Neural Networks.

Self-Ablating Transformers: More Interpretability, Less Sparsity Gradient Routing: Masking Gradients to Localize Computation in Neural Networks

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.520989Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 49852d33-27fc-47d1-98ce-2b0c8598d3dd · outbound

This paper cites Mavor-Parker, Aengus Lynch, Stefan Heimersheim, and Adri \`a Garriga-Alonso.

Self-Ablating Transformers: More Interpretability, Less Sparsity Mavor-Parker, Aengus Lynch, Stefan Heimersheim, and Adri \`a Garriga-Alonso

Reference 5

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:43:46.081343Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation ea230a7e-0b3e-4bd8-a0d6-707adb01d225 · outbound

This paper cites TinyStories: How Small Can Language Models Be and Still Speak Coherent English?.

Self-Ablating Transformers: More Interpretability, Less Sparsity TinyStories: How Small Can Language Models Be and Still Speak Coherent English?

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.529143Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 28990087-6b1a-4341-9be6-fa54a9325584 · outbound

This paper cites Neuron to Graph: Interpreting Language Model Neurons at Scale.

Self-Ablating Transformers: More Interpretability, Less Sparsity Neuron to Graph: Interpreting Language Model Neurons at Scale

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.533558Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3655251c-b0ab-4b08-81bb-5cec5437e60f · outbound

This paper cites N2G: A Scalable Approach for Quantifying Interpretable Neuron Representations in Large Language Models.

Self-Ablating Transformers: More Interpretability, Less Sparsity N2G: A Scalable Approach for Quantifying Interpretable Neuron Representations in Large Language Models

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.537514Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c9449e0d-058e-4163-821a-0bfa8d659aa5 · outbound

This paper cites Stabilizing the Lottery Ticket Hypothesis.

Self-Ablating Transformers: More Interpretability, Less Sparsity Stabilizing the Lottery Ticket Hypothesis

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.541038Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d137550d-efa3-4484-a9b5-14f9bc90c2d4 · outbound

This paper cites The Pile: An 800GB Dataset of Diverse Text for Language Modeling.

Self-Ablating Transformers: More Interpretability, Less Sparsity The Pile: An 800GB Dataset of Diverse Text for Language Modeling

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.544672Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 22536831-2b59-4e7e-bb65-266c137f10de · outbound

This paper cites Scaling and evaluating sparse autoencoders.

Self-Ablating Transformers: More Interpretability, Less Sparsity Scaling and evaluating sparse autoencoders

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.548149Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6e7901d4-411d-4bd2-86d9-c2dc3eaf38f7 · outbound

This paper cites Gemini: A Family of Highly Capable Multimodal Models.

Self-Ablating Transformers: More Interpretability, Less Sparsity Gemini: A Family of Highly Capable Multimodal Models

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.551102Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ed9e84da-f5f8-4d37-9d15-a0b2c8c5f419 · outbound

This paper cites FrontierMath: A Benchmark for Evaluating Advanced Mathematical Reasoning in AI.

Self-Ablating Transformers: More Interpretability, Less Sparsity FrontierMath: A Benchmark for Evaluating Advanced Mathematical Reasoning in AI

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.553635Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2ee220c7-6eb6-4ed7-acbb-c9720dd2504f · outbound

This paper cites Openwebtext corpus.

Self-Ablating Transformers: More Interpretability, Less Sparsity Openwebtext corpus

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.556523Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8cad3ac6-af2f-494a-8f5f-5cea8ad1eb35 · outbound

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

Self-Ablating Transformers: More Interpretability, Less Sparsity Sparse autoencoders find highly interpretable features in language models

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.559097Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c4327362-cad3-4d04-9892-6e50a7664e36 · outbound

This paper cites Two sparsities are better than one: unlocking the performance benefits of sparse–sparse networks.

Self-Ablating Transformers: More Interpretability, Less Sparsity Two sparsities are better than one: unlocking the performance benefits of sparse–sparse networks

Reference 16

Resolution: verified exact; doi, observed 2026-08-16T04:43:45.694494Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 269f3908-1896-4f84-9965-384c58b3db09 · outbound

This paper cites an unresolved cited work.

Self-Ablating Transformers: More Interpretability, Less Sparsity Unresolved cited work

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.565343Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ba88fd9f-15d1-471c-b721-c4f32dd2e5d3 · outbound

This paper cites Lazzaro, S.

Self-Ablating Transformers: More Interpretability, Less Sparsity Lazzaro, S

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:43:46.052630Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 6085f51d-8a80-4faa-9d84-051192654ce1 · outbound

This paper cites Lecun, L.

Self-Ablating Transformers: More Interpretability, Less Sparsity Lecun, L

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.571944Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1c4f1516-fa8e-4ad7-a72f-69555bf19ab9 · outbound

This paper cites Deep learning.

Self-Ablating Transformers: More Interpretability, Less Sparsity Deep learning

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.575516Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2cc3b821-cda7-435b-b31a-ca6d523c561c · outbound

This paper cites The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning.

Self-Ablating Transformers: More Interpretability, Less Sparsity The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.579184Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6dd5c797-503a-474a-8f15-643e781b03d7 · outbound

This paper cites KAN: Kolmogorov-Arnold Networks.

Self-Ablating Transformers: More Interpretability, Less Sparsity KAN: Kolmogorov-Arnold Networks

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.583081Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 82319611-8522-4537-8833-a45a97063fe5 · outbound

This paper cites Majani, Ruth Erlanson, and Yaser Abu-Mostafa.

Self-Ablating Transformers: More Interpretability, Less Sparsity Majani, Ruth Erlanson, and Yaser Abu-Mostafa

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:43:46.040037Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation b8c3debc-0212-48a1-b138-862bd6c09a29 · outbound

This paper cites Transformer circuit evaluation metrics are not robust.

Self-Ablating Transformers: More Interpretability, Less Sparsity Transformer circuit evaluation metrics are not robust

Reference 24

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:43:46.028393Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 72949077-1963-4077-8d90-a55c1fad2c37 · outbound

This paper cites GPT-4o mini: advancing cost-efficient intelligence , 2024.

Self-Ablating Transformers: More Interpretability, Less Sparsity GPT-4o mini: advancing cost-efficient intelligence , 2024

Reference 25

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:43:46.016687Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation a50be9b8-c0e8-4689-8843-2a24132ba8e9 · outbound

This paper cites GPT-4 Technical Report.

Self-Ablating Transformers: More Interpretability, Less Sparsity GPT-4 Technical Report

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.597501Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9965ed0e-3f4f-4d70-ba14-cadc8ab51d42 · outbound

This paper cites why should i trust you?.

Self-Ablating Transformers: More Interpretability, Less Sparsity why should i trust you?

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.601541Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3b5abdd8-36a2-4c5f-b059-bf60fb302c5c · outbound

This paper cites Stop Explaining Black Box Machine Learning Models for High Stakes Decisions and Use Interpretable Models Instead.

Self-Ablating Transformers: More Interpretability, Less Sparsity Stop Explaining Black Box Machine Learning Models for High Stakes Decisions and Use Interpretable Models Instead

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.604891Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 98a2693c-06e7-46f3-b96a-6520b5e8bfad · outbound

This paper cites SAEBench: A Comprehensive Benchmark for Sparse Autoencoders - Dec 2024 , January 2025.

Self-Ablating Transformers: More Interpretability, Less Sparsity SAEBench: A Comprehensive Benchmark for Sparse Autoencoders - Dec 2024 , January 2025

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:43:46.004687Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 8ecd6819-0c28-4f33-8f3a-c9c8fc1fea21 · outbound

This paper cites DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter.

Self-Ablating Transformers: More Interpretability, Less Sparsity DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.612842Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7003ab81-db8a-4725-880b-0d51a484d87d · outbound

This paper cites The neural lasso: Local linear sparsity for interpretable explanations.

Self-Ablating Transformers: More Interpretability, Less Sparsity The neural lasso: Local linear sparsity for interpretable explanations

Reference 31

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:43:45.992816Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 369188da-d88d-4bae-b066-337bac871d4f · outbound

This paper cites Dropout: A simple way to prevent neural networks from overfitting.

Self-Ablating Transformers: More Interpretability, Less Sparsity Dropout: A simple way to prevent neural networks from overfitting

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.618279Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 855a7c8c-0a35-4b55-88dd-0cbe1ec9fbd5 · outbound

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

Self-Ablating Transformers: More Interpretability, Less Sparsity Codebook Features: Sparse and Discrete Interpretability for Neural Networks

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.621423Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 78e7904b-0e24-4f5a-ab81-63482c1df7ab · outbound

This paper cites Attention is all you need.

Self-Ablating Transformers: More Interpretability, Less Sparsity Attention is all you need

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.624315Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 342cadae-d6d4-4391-a89c-025172654035 · outbound

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

Self-Ablating Transformers: More Interpretability, Less Sparsity Interpretability in the wild: a circuit for indirect object identification in GPT -2 small

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.628392Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c37afdf1-3b83-4e62-b18c-2dae14fe1f99 · outbound

This paper cites MMLU -pro: A more robust and challenging multi-task language understanding benchmark.

Self-Ablating Transformers: More Interpretability, Less Sparsity MMLU -pro: A more robust and challenging multi-task language understanding benchmark

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.633209Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 18097027-a1fd-4a2c-bc16-d81cbbec50cc · outbound

This paper cites Understanding deep learning requires rethinking generalization.

Self-Ablating Transformers: More Interpretability, Less Sparsity Understanding deep learning requires rethinking generalization

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.637041Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 578d8e96-1a14-493f-a267-199101320d09 · outbound

This paper cites Towards best practices of activation patching in language models: Metrics and methods.

Self-Ablating Transformers: More Interpretability, Less Sparsity Towards best practices of activation patching in language models: Metrics and methods

Reference 38

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:43:45.956445Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 70200a96-f2ef-4ca6-be3f-747401284603 · outbound

This paper cites @esa (Ref.

Self-Ablating Transformers: More Interpretability, Less Sparsity @esa (Ref

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.644152Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 60a58f32-d57d-4753-9079-9dd0f9ee2460 · outbound

This paper cites an unresolved cited work.

Self-Ablating Transformers: More Interpretability, Less Sparsity Unresolved cited work

Reference 40

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.648135Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 12c73708-74c6-4cb9-8ced-61dd33e6d50a · outbound

This paper cites an unresolved cited work.

Self-Ablating Transformers: More Interpretability, Less Sparsity Unresolved cited work

Reference 41

Resolution: unresolved; no resolver link, observed 2026-08-16T04:43:45.651719Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

No inbound Pith citation observations are available.