Pith. sign in

Paper Citation Record · LEDGER

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers

As of 18 August 2026, this Paper Citation Record lists 46 of 46 outbound references and 1 inbound Pith citation observation for arXiv:2411.08745.

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

pith.paper-citation-record.v1
2411.08745 v4

Coverage vector

measured 46 of 46 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-12T21:29:17.161324Z

measured 47 of 47 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-17T06:30:58.91139+00:00

measured 1 of 1 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-06T13:48:09.519547Z

measured 0 of 1 external citation measurements

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

Source: pith, observed 2026-08-06T13:48:13.263301Z

Reference resolution

46 of 46 outbound references displayed

  • verified exact1
  • verified fuzzy0
  • unresolved45
  • parse uncertain0
  • malformed identifier0
  • metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 6f274894-38fc-46c5-bd94-6b84b32af132 · outbound

This paper cites online" 'onlinestring :=.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers online" 'onlinestring :=

Reference 1

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:16.996273Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:16.996273Z digest=sha256:be7c67ec107bca5086a32dc43e1742fdbf1b2d19bd822eefe70741425baf3afc

Observation 1528abff-1f88-43b2-abe1-1df493add53a · outbound

This paper cites write newline.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers write newline

Reference 2

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.000489Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.000489Z digest=sha256:bd74381d7ba1aedd150d00909030e80a951c91c73917e580016027dd579aaf11

Observation 096b59bb-8c68-4574-aa45-fd9d7f63ec7c · outbound

This paper cites an unresolved cited work.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Unresolved cited work

Reference 3

Resolution
unresolved
raw_fallback, observed 2026-08-12T21:29:17.575277Z

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.

source=arxiv_source observed=2026-08-12T21:29:17.004605Z digest=sha256:aa3812a0797267e72edb0eb401afe82db5a618a46e00055590a281adde663d20

Observation 9e143d08-eee1-4a30-ae59-847d66654d11 · outbound

This paper cites Aya 23: Open Weight Releases to Further Multilingual Progress.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Aya 23: Open Weight Releases to Further Multilingual Progress

Reference 4

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.007784Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.007784Z digest=sha256:b7f23131a4beaf872bb64b8afd25ed556162185d122daa96e49e8f61159be5f8

Observation c04d130c-f253-4649-9684-58a9b403ed0b · outbound

This paper cites Qwen Technical Report.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Qwen Technical Report

Reference 5

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.011443Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.011443Z digest=sha256:7d65c49c92ad5f24b670b9ec8215f7cde01632c573418e3c872bcfa4ac7c37c1

Observation cfabfdde-8b3a-439f-a0c4-f03a282e989e · outbound

This paper cites an unresolved cited work.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Unresolved cited work

Reference 6

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.015265Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.015265Z digest=sha256:6ec67fafcdffa8f17f8ba093d11c254680acc382aee0101eeda1e3bf880583df

Observation 0d3b644c-772e-4483-be71-f65d615a31ef · outbound

This paper cites One Mind, Many Tongues: A Deep Dive into Language-Agnostic Knowledge Neurons in Large Language Models.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers One Mind, Many Tongues: A Deep Dive into Language-Agnostic Knowledge Neurons in Large Language Models

Reference 7

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.019011Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.019011Z digest=sha256:83f7bb722f90026b22766f7d8734c2175e9da3e8db10bf70052a31ce4ff0bbfe

Observation 976f3c69-9f3b-4bc0-8446-706713e09208 · outbound

This paper cites SelfIE: Self-Interpretation of Large Language Model Embeddings.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers SelfIE: Self-Interpretation of Large Language Model Embeddings

Reference 8

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.023640Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.023640Z digest=sha256:c1d1d047715bc706b92c3e309314a480d0f593dc8a1a1ac01366136d348836e2

Observation e2a026fc-1770-4500-8a56-ab0091606cee · outbound

This paper cites an unresolved cited work.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Unresolved cited work

Reference 9

Resolution
unresolved
raw_fallback, observed 2026-08-12T21:29:17.558781Z

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.

source=arxiv_source observed=2026-08-12T21:29:17.028278Z digest=sha256:2f8803c2e03340a33d05f29201b0ae9ef4dbce636b8722fe0ebd4a4366cd554b

Observation d1887500-9d19-492c-9c11-bf1966b22472 · outbound

This paper cites Zero-shot cross-lingual transfer in instruction tuning of large language models.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Zero-shot cross-lingual transfer in instruction tuning of large language models

Reference 10

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.032091Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.032091Z digest=sha256:cecc4c82a64915cca3b6f44ccdf5134a0267103520eeae85c0a06a1ec5abb932

Observation b5262f87-f15a-4d36-9916-c70c88bad62a · outbound

This paper cites What does it mean to be language-agnostic? Probing multilingual sentence encoders for typological properties.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers What does it mean to be language-agnostic? Probing multilingual sentence encoders for typological properties

Reference 11

Resolution
verified exact
local_arxiv, observed 2026-08-12T21:29:17.430329Z

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.

source=arxiv_source observed=2026-08-12T21:29:17.036091Z digest=sha256:7d47f9b917c4e281734c12b6bc912f0fec707f3bd06c64014cd584c2cdb462e6

Observation cc8e8ed4-7624-4380-90d0-80a5963536b5 · outbound

This paper cites Unsupervised Cross-lingual Representation Learning at Scale.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Unsupervised Cross-lingual Representation Learning at Scale

Reference 12

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.040064Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.040064Z digest=sha256:897273f82360f63770c5e6fbc07d93eb5d3085696ca630d102c0f6d1e92f5e04

Observation afa0696c-b5fe-44d9-b2a7-6cd3a08b3d19 · outbound

This paper cites an unresolved cited work.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Unresolved cited work

Reference 13

Resolution
unresolved
raw_fallback, observed 2026-08-12T21:29:17.547309Z

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.

source=arxiv_source observed=2026-08-12T21:29:17.043739Z digest=sha256:e7e08821137bdf555b61f77d30b5b0fc77c75ec9d019b50b6c7f5671b63409bb

Observation b67494f5-e062-4815-86b6-4ddc26fa48e2 · outbound

This paper cites BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

Reference 14

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.047299Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.047299Z digest=sha256:649a4b8d73aeb46feb19678a524e6d81dff0f5a9a80475a00ff78dc38c767bcf

Observation e81dea49-67f9-4c0a-84f1-92123bcb573d · outbound

This paper cites The Llama 3 Herd of Models.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers The Llama 3 Herd of Models

Reference 15

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.051155Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.051155Z digest=sha256:3c208db1ffad95fe70e2e3c8c3674918a36ed5938502cac6f0a90dae7f988818

Observation 5d78b88e-a629-4b19-affc-ad9c8d34693f · outbound

This paper cites NNsight and NDIF: Democratizing Access to Open-Weight Foundation Model Internals.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers NNsight and NDIF: Democratizing Access to Open-Weight Foundation Model Internals

Reference 16

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.054612Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.054612Z digest=sha256:05f28205224813b193c0117ce80d69dc2608881a4216c1f364733b822bdd8d68

Observation 88abdd4a-b217-4a79-9792-505e19b4f72c · outbound

This paper cites an unresolved cited work.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Unresolved cited work

Reference 17

Resolution
unresolved
raw_fallback, observed 2026-08-12T21:29:17.536952Z

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.

source=arxiv_source observed=2026-08-12T21:29:17.058175Z digest=sha256:358e1c32b23052b4e36c5f7cb6619108205a85532953192210686f1921b1ce7f

Observation 9acc1445-2cd0-4047-9f4a-2d6c04bfc610 · outbound

This paper cites Patchscopes: A Unifying Framework for Inspecting Hidden Representations of Language Models.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Patchscopes: A Unifying Framework for Inspecting Hidden Representations of Language Models

Reference 18

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.061584Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.061584Z digest=sha256:699e05315ca265fd3c75d694593a90d1d6f2abe92cf2c64a7179164c1cb250ae

Observation 6923915f-a992-4675-9fb3-c9a9cc6b50d7 · outbound

This paper cites Mistral 7B.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Mistral 7B

Reference 19

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.065753Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.065753Z digest=sha256:c5261f6ea1d1cab026679b70c4ffd5f57e366440c4a2dda3d227f4f96291af87

Observation a9fc4b30-04b2-4085-9193-6378f20f8b78 · outbound

This paper cites AtP*: An efficient and scalable method for localizing LLM behaviour to components.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers AtP*: An efficient and scalable method for localizing LLM behaviour to components

Reference 20

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.070133Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.070133Z digest=sha256:cec2a289576383bd9f0fb097f2a035181df70dc8a93c1dc1390bcfaec417974a

Observation 8a6d1d1a-aaab-4737-9f83-cf1f3894c547 · outbound

This paper cites an unresolved cited work.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Unresolved cited work

Reference 21

Resolution
unresolved
raw_fallback, observed 2026-08-12T21:29:17.527626Z

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.

source=arxiv_source observed=2026-08-12T21:29:17.073952Z digest=sha256:73c4a195b021f6e4650e8168f413ec9f5ac6381735da8c4ff6062843e1fded31

Observation 6fa3ee6e-e11d-4038-9d75-5f091ee2a964 · outbound

This paper cites On the Language Neutrality of Pre-trained Multilingual Representations.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers On the Language Neutrality of Pre-trained Multilingual Representations

Reference 22

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.077352Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.077352Z digest=sha256:684eb3efb53a3fbe67637001f37e3e335d0bfc40a607c71833f328c2c42b6ee5

Observation b2ee9281-a7d5-4dc0-abff-b4b2059301ca · outbound

This paper cites an unresolved cited work.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Unresolved cited work

Reference 23

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.080882Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.080882Z digest=sha256:3127ff42d5fbb5feed36876e61cfadfb42e80ccd23f046caba45e24c77e317e0

Observation 528c63ec-5826-4090-b246-15871039c59b · outbound

This paper cites an unresolved cited work.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Unresolved cited work

Reference 24

Resolution
unresolved
raw_fallback, observed 2026-08-12T21:29:17.512207Z

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.

source=arxiv_source observed=2026-08-12T21:29:17.084019Z digest=sha256:151744460b4e86e408e7eeb89c59285e771567027c850fe3ad5df8931bb9a5c8

Observation 9ceaf065-3bb2-40d6-ae57-342ec1682b0c · outbound

This paper cites an unresolved cited work.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Unresolved cited work

Reference 25

Resolution
unresolved
raw_fallback, observed 2026-08-12T21:29:17.502363Z

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.

source=arxiv_source observed=2026-08-12T21:29:17.087562Z digest=sha256:c86b259fdbc30266a49c0874fee7d037de0ff55fd716db7f88b4dd48d7eb196d

Observation 8c8dcb7f-c3d7-49e2-97ff-29cef7834aff · outbound

This paper cites an unresolved cited work.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Unresolved cited work

Reference 26

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.090875Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.090875Z digest=sha256:6ddf9e7f489966f2a5d2d10f1faaa9d504a7ee591975478fadddd365a172aa39

Observation 7a0d342d-0217-4e9e-9fe2-d5343d27f3bb · outbound

This paper cites Whose Opinions Do Language Models Reflect?.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Whose Opinions Do Language Models Reflect?

Reference 27

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.094332Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.094332Z digest=sha256:e22f7290e29b9ff31a2b8aa196d48c4fcec88b85c76eb896aeb88d4e6dc14739

Observation 217c9daf-61dc-4761-a24e-42a9b27be025 · outbound

This paper cites How multilingual is Multilingual BERT?.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers How multilingual is Multilingual BERT?

Reference 28

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.097690Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.097690Z digest=sha256:fb99beaca2559261d31ff143cbe584f30e3df557afa7176c74ae22a2d3263de6

Observation bb0debbd-da69-40de-8eec-a6c42e1aac2e · outbound

This paper cites an unresolved cited work.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Unresolved cited work

Reference 29

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.101121Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.101121Z digest=sha256:6c7c4a47bf04c8e8a8dda725efeafa6183840692922508ead057a999b7118c65

Observation 6027bf2f-5e00-47bb-9988-ad6b270fa4d8 · outbound

This paper cites Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks

Reference 30

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.104521Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.104521Z digest=sha256:00b9c1926ed90e377b708b60d7b88aee6f714e297d64f71694919bbb6e9088ea

Observation d98b108f-2155-46e8-8ad8-290d71d45d64 · outbound

This paper cites Making Monolingual Sentence Embeddings Multilingual using Knowledge Distillation.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Making Monolingual Sentence Embeddings Multilingual using Knowledge Distillation

Reference 31

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.107909Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.107909Z digest=sha256:0c8604b867ab36c2828d1cfa1b48a18c7e44a71249b6f0127759677fe117405e

Observation b6207394-958b-4ac5-af73-291a698b511e · outbound

This paper cites Multilingual Instruction Tuning With Just a Pinch of Multilinguality.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Multilingual Instruction Tuning With Just a Pinch of Multilinguality

Reference 32

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.111411Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.111411Z digest=sha256:246601c3e324721a415f59659e33f706eba9f77b4e91ba61d195ed5cdbbc8b86

Observation 61e9f823-2497-46e5-a73d-1c602c8c9e55 · outbound

This paper cites Language Models are Multilingual Chain-of-Thought Reasoners.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Language Models are Multilingual Chain-of-Thought Reasoners

Reference 33

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.115182Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.115182Z digest=sha256:73e7a714407faaa46a9500581f4e4594a0924d8b6035388c19ea12c1825eb45f

Observation 4f887ca2-122a-45ea-a8e6-10af7d9f89b1 · outbound

This paper cites MPNet: Masked and Permuted Pre-training for Language Understanding.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers MPNet: Masked and Permuted Pre-training for Language Understanding

Reference 34

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.118942Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.118942Z digest=sha256:438da0e6ac1be36def930deebfeed2c1d217af4493c12731a18d97368cebe4c1

Observation 8bb7185c-2916-4ec6-bedf-33186b4be253 · outbound

This paper cites Same Neurons, Different Languages: Probing Morphosyntax in Multilingual Pre-trained Models.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Same Neurons, Different Languages: Probing Morphosyntax in Multilingual Pre-trained Models

Reference 35

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.122567Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.122567Z digest=sha256:4fbaebf003247219a421819253ceb0991485cd7b5a5c2a9b2fb9b62f826587fa

Observation 66c4bfd7-0757-47a1-87a2-2812b678bcef · outbound

This paper cites Language-Specific Neurons: The Key to Multilingual Capabilities in Large Language Models.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Language-Specific Neurons: The Key to Multilingual Capabilities in Large Language Models

Reference 36

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.125694Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.125694Z digest=sha256:34e2aaaf6c446f74462f84ab4bbea7ef90f6e7f61615bcf4471927432014c14d

Observation b94788f5-313b-4e56-b506-0ae990f94510 · outbound

This paper cites Gemma 2: Improving Open Language Models at a Practical Size.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Gemma 2: Improving Open Language Models at a Practical Size

Reference 37

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.128852Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.128852Z digest=sha256:a82c4a379477cf8fff5fb4290cdb81e30e56ac85e636515a1406c50067b848c7

Observation 29839ecc-a41b-4a83-bb88-b16a8d986ebe · outbound

This paper cites Function Vectors in Large Language Models.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Function Vectors in Large Language Models

Reference 38

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.133503Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.133503Z digest=sha256:42dce556ba71c60ee1eed0980826cadcaf38620facf8113cd62aa4b90621f4c9

Observation b551521d-07a7-4009-b9b5-fa1b1fb447d6 · outbound

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

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Llama 2: Open Foundation and Fine-Tuned Chat Models

Reference 39

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.136990Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.136990Z digest=sha256:a08dd75acbb3b96911936d61d329e007fa7457fa680419823de346fc14bd22c4

Observation ca9d035e-40f0-4b3b-b626-2e6be933fcc7 · outbound

This paper cites Look Before You Leap: A Universal Emergent Decomposition of Retrieval Tasks in Language Models.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Look Before You Leap: A Universal Emergent Decomposition of Retrieval Tasks in Language Models

Reference 40

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.140227Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.140227Z digest=sha256:563ea870fe9d56de172076c63dd3fad89b776e6297b2e9a4ed6e363daa8fe385

Observation 9f57ee63-302b-4aa0-b1b9-c714af47f324 · outbound

This paper cites an unresolved cited work.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Unresolved cited work

Reference 41

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.143707Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.143707Z digest=sha256:cc1006cc5d9f20a778d5632e512ccd73604349b9a82896f3f91524884a497b34

Observation b59c0a9e-2125-4191-8298-499b2394c980 · outbound

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

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Do Llamas Work in English? On the Latent Language of Multilingual Transformers

Reference 42

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.147041Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.147041Z digest=sha256:67ecd0b47abe12c18c97d6dc6ba26c4d20371d751ff2bd57ffa13b8ea592e5bd

Observation babc7e52-97a7-4b0b-b42b-bf5e99ed5221 · outbound

This paper cites Emerging Cross-lingual Structure in Pretrained Language Models.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Emerging Cross-lingual Structure in Pretrained Language Models

Reference 43

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.150817Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.150817Z digest=sha256:123c495b0bf0e850dbf21bec00341e045391ad7a5e3d555614cd558805763f8a

Observation e909ceb4-a2a9-4454-9b05-0fe862d6bf3a · outbound

This paper cites The Semantic Hub Hypothesis: Language Models Share Semantic Representations Across Languages and Modalities.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers The Semantic Hub Hypothesis: Language Models Share Semantic Representations Across Languages and Modalities

Reference 44

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.154407Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.154407Z digest=sha256:ee9ee9290bfc5eaddb66e84fbbccd467c747cf59c4ea76715a090992ddd9fb3b

Observation 8dc1e85c-caa5-4ddf-97cc-41d1a84d793b · outbound

This paper cites an unresolved cited work.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Unresolved cited work

Reference 45

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.157826Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.157826Z digest=sha256:e073540114f1d2afef35d20e09870a58342abcde75ad5c5153190c5dd07cb511

Observation c077bdc2-d2d1-4bca-8d6d-6f351615a816 · outbound

This paper cites Converging to a Lingua Franca: Evolution of Linguistic Regions and Semantics Alignment in Multilingual Large Language Models.

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers Converging to a Lingua Franca: Evolution of Linguistic Regions and Semantics Alignment in Multilingual Large Language Models

Reference 46

Resolution
unresolved
no resolver link, observed 2026-08-12T21:29:17.161324Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-12T21:29:17.161324Z digest=sha256:466ba1de9400f43a485341922f496815bd2bb825f6dc5669eff6134171113133

Pith citing papers

Observation b1a71b94-180c-4912-9ae1-3e45665a914a · inbound

What Language(s) Does Aya-23 Think In? How Multilinguality Affects Internal Language Representations cites this paper.

What Language(s) Does Aya-23 Think In? How Multilinguality Affects Internal Language Representations Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers

Reference 2024

Resolution
metadata mismatch
local_arxiv, observed 2026-08-06T13:48:13.418180Z

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.

source=pdf_text observed=2026-08-06T13:48:09.519547Z digest=sha256:4347fb5f240170faf977c81f00dc06989511c2aee62e1c7f7c10a984cf89484b