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

arXiv: https://arxiv.org/abs/2411.08745 · HTML: https://mlsanity.com/paper/2411.08745/

## Key facts

- Cited by 44 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning.
- Builds on 33 of its 50 references that are in the map.
- First posted on arXiv 13 Nov 2024.

## Placed in

- LLM reasoning / [Reasoning benchmarks and evaluation](https://mlsanity.com/lr/reasoning-benchmarks-and-evaluation/)
- Mechanistic interpretability / [Knowledge and representations inside LLMs](https://mlsanity.com/mechinterp/knowledge-and-representations-inside-llms/)

## Builds on (most cited first, 30 of 33)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [Qwen Technical Report](https://mlsanity.com/paper/2309.16609/) (2023)
- [Mistral 7B](https://mlsanity.com/paper/2310.06825/) (2023)
- [Gemma 2: Improving Open Language Models at a Practical Size](https://mlsanity.com/paper/2408.00118/) (2024)
- [MPNet: Masked and Permuted Pre-training for Language Understanding](https://mlsanity.com/paper/2004.09297/) (2020)
- [Whose Opinions Do Language Models Reflect?](https://mlsanity.com/paper/2303.17548/) (2023)
- [How multilingual is Multilingual BERT?](https://mlsanity.com/paper/1906.01502/) (2019)
- [Making Monolingual Sentence Embeddings Multilingual using Knowledge Distillation](https://mlsanity.com/paper/2004.09813/) (2020)
- [Language Models are Multilingual Chain-of-Thought Reasoners](https://mlsanity.com/paper/2210.03057/) (2022)
- [Function Vectors in Large Language Models](https://mlsanity.com/paper/2310.15213/) (2023)
- [Do Llamas Work in English? On the Latent Language of Multilingual Transformers](https://mlsanity.com/paper/2402.10588/) (2024)
- [Patchscopes: A Unifying Framework for Inspecting Hidden Representations of Language Models](https://mlsanity.com/paper/2401.06102/) (2024)
- [Language-Specific Neurons: The Key to Multilingual Capabilities in Large Language Models](https://mlsanity.com/paper/2402.16438/) (2024)
- [Aya 23: Open Weight Releases to Further Multilingual Progress](https://mlsanity.com/paper/2405.15032/) (2024)
- [AtP*: An efficient and scalable method for localizing LLM behaviour to components](https://mlsanity.com/paper/2403.00745/) (2024)
- [Finding Universal Grammatical Relations in Multilingual BERT](https://mlsanity.com/paper/2005.04511/) (2020)
- [On the Language Neutrality of Pre-trained Multilingual Representations](https://mlsanity.com/paper/2004.05160/) (2020)
- [Multilingual Instruction Tuning With Just a Pinch of Multilinguality](https://mlsanity.com/paper/2401.01854/) (2024)
- [The Semantic Hub Hypothesis: Language Models Share Semantic Representations Across Languages and Modalities](https://mlsanity.com/paper/2411.04986/) (2024)
- [SelfIE: Self-Interpretation of Large Language Model Embeddings](https://mlsanity.com/paper/2403.10949/) (2024)
- [NNsight and NDIF: Democratizing Access to Open-Weight Foundation Model Internals](https://mlsanity.com/paper/2407.14561/) (2024)
- [What does it mean to be language-agnostic? Probing multilingual sentence encoders for typological properties](https://mlsanity.com/paper/2009.12862/) (2020)
- [Converging to a Lingua Franca: Evolution of Linguistic Regions and Semantics Alignment in Multilingual Large Language Models](https://mlsanity.com/paper/2410.11718/) (2024)
- [How Do Multilingual Language Models Remember Facts?](https://mlsanity.com/paper/2410.14387/) (2024)
- [Semeval-2022 Task 1: CODWOE -- Comparing Dictionaries and Word Embeddings](https://mlsanity.com/paper/2205.13858/) (2022)
- [Look Before You Leap: A Universal Emergent Decomposition of Retrieval Tasks in Language Models](https://mlsanity.com/paper/2312.10091/) (2023)

## Built on by (most cited first, 30 of 44)

- [The Dual-Route Model of Induction](https://mlsanity.com/paper/2504.03022/) (2025)
- [Localized Cultural Knowledge is Conserved and Controllable in Large Language Models](https://mlsanity.com/paper/2504.10191/) (2025)
- [RomanLens: The Role Of Latent Romanization In Multilinguality In LLMs](https://mlsanity.com/paper/2502.07424/) (2025)
- [Can you map it to English? The Role of Cross-Lingual Alignment in Multilingual Performance of LLMs](https://mlsanity.com/paper/2504.09378/) (2025)
- [Internal Chain-of-Thought: Empirical Evidence for Layer-wise Subtask Scheduling in LLMs](https://mlsanity.com/paper/2505.14530/) (2025)
- [LLMs Process Lists With General Filter Heads](https://mlsanity.com/paper/2510.26784/) (2025)
- [Sparse Feature Coactivation Reveals Causal Semantic Modules in Large Language Models](https://mlsanity.com/paper/2506.18141/) (2025)
- [Knowledge Localization in Mixture-of-Experts LLMs Using Cross-Lingual Inconsistency](https://mlsanity.com/paper/2603.17102/) (2026)
- [What Language(s) Does Aya-23 Think In? How Multilinguality Affects Internal Language Representations](https://mlsanity.com/paper/2507.20279/) (2025)
- [When Meanings Meet: Investigating the Emergence and Quality of Shared Concept Spaces during Multilingual Language Model Training](https://mlsanity.com/paper/2601.22851/) (2026)
- [CoCo-CoLa: Evaluating and Improving Language Adherence in Multilingual LLMs](https://mlsanity.com/paper/2502.12476/) (2025)
- [On the Entity-Level Alignment in Crosslingual Consistency](https://mlsanity.com/paper/2510.10280/) (2025)
- [Latent Concept Disentanglement in Transformer-based Language Models](https://mlsanity.com/paper/2506.16975/) (2025)
- [Interpreting Language Models Through Concept Descriptions: A Survey](https://mlsanity.com/paper/2510.01048/) (2025)
- [Counting Circuits: Mechanistic Interpretability of Visual Reasoning in Large Vision-Language Models](https://mlsanity.com/paper/2603.18523/) (2026)
- [In-Context Learning Operates as Concept Subspace Learning](https://mlsanity.com/paper/2605.18830/) (2026)
- [XToM: Exploring the Multilingual Theory of Mind for Large Language Models](https://mlsanity.com/paper/2506.02461/) (2025)
- [YaPO: Learnable Sparse Activation Steering Vectors for Domain Adaptation](https://mlsanity.com/paper/2601.08441/) (2026)
- [Layer-Targeted Multilingual Knowledge Erasure in Large Language Models](https://mlsanity.com/paper/2602.22562/) (2026)
- [How LLMs Are Persuaded: A Few Attention Heads, Rerouted](https://mlsanity.com/paper/2605.09314/) (2026)
- [Low-Resource Safety Failures Are Action Failures, Not Representation Failures](https://mlsanity.com/paper/2606.01196/) (2026)
- [Parallel Universes, Parallel Languages: A Comprehensive Study on LLM-based Multilingual Counterfactual Example Generation](https://mlsanity.com/paper/2601.00263/) (2026)
- [LangFIR: Discovering Sparse Language-Specific Features from Monolingual Data for Language Steering](https://mlsanity.com/paper/2604.03532/) (2026)
- [Copy First, Translate Later: Interpreting Translation Dynamics in Multilingual Pretraining](https://mlsanity.com/paper/2604.17633/) (2026)
- [Beyond Language: Format-Agnostic Reasoning Subspaces in Large Language Models](https://mlsanity.com/paper/2605.09496/) (2026)
- [Shared Doubt: Zero-Shot Cross-Lingual Confidence Estimation for Language Models](https://mlsanity.com/paper/2605.31220/) (2026)
- [Test, then Route: How Language Models Execute In-Context Conditional Rules Across Models and Languages](https://mlsanity.com/paper/2608.04183/) (2026)
- [SteerEval: Inference-time Interventions Strengthen Multilingual Generalization in Neural Summarization Metrics](https://mlsanity.com/paper/2601.15809/) (2026)
- [Identifying Crucial Attention Heads for Multilingual Language Models: Retrieval and Retrieval-Transition Heads](https://mlsanity.com/paper/2602.22453/) (2026)
- [CoSToM:Causal-oriented Steering for Intrinsic Theory-of-Mind Alignment in Large Language Models](https://mlsanity.com/paper/2604.10031/) (2026)

## Abstract (arXiv)

A central question in multilingual language modeling is whether large language models (LLMs) develop a universal concept representation, disentangled from specific languages. In this paper, we address this question by analyzing latent representations (latents) during a word-translation task in transformer-based LLMs. We strategically extract latents from a source translation prompt and insert them into the forward pass on a target translation prompt. By doing so, we find that the output language is encoded in the latent at an earlier layer than the concept to be translated. Building on this insight, we conduct two key experiments. First, we demonstrate that we can change the concept without changing the language and vice versa through activation patching alone. Second, we show that patching with the mean representation of a concept across different languages does not affect the models' ability to translate it, but instead improves it. Finally, we generalize to multi-token generation and demonstrate that the model can generate natural language description of those mean representations. Our results provide evidence for the existence of language-agnostic concept representations within the investigated models.