# On the Similarity of Circuits across Languages: a Case Study on the Subject-verb Agreement Task

arXiv: https://arxiv.org/abs/2410.06496 · HTML: https://mlsanity.com/paper/2410.06496/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 38 references that are in the map.
- First posted on arXiv 9 Oct 2024.

## Builds on (most cited first, 20 of 20)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [PaLM: Scaling Language Modeling with Pathways](https://mlsanity.com/paper/2204.02311/) (2022)
- [Locating and Editing Factual Associations in GPT](https://mlsanity.com/paper/2202.05262/) (2022)
- [Inference-Time Intervention: Eliciting Truthful Answers from a Language Model](https://mlsanity.com/paper/2306.03341/) (2023)
- [Interpretability in the Wild: a Circuit for Indirect Object Identification in GPT-2 small](https://mlsanity.com/paper/2211.00593/) (2022)
- [Probing Classifiers: Promises, Shortcomings, and Advances](https://mlsanity.com/paper/2102.12452/) (2021)
- [Dissecting Recall of Factual Associations in Auto-Regressive Language Models](https://mlsanity.com/paper/2304.14767/) (2023)
- [Causal Abstractions of Neural Networks](https://mlsanity.com/paper/2106.02997/) (2021)
- [Assessing the Ability of LSTMs to Learn Syntax-Sensitive Dependencies](https://mlsanity.com/paper/1611.01368/) (2016)
- [Do Llamas Work in English? On the Latent Language of Multilingual Transformers](https://mlsanity.com/paper/2402.10588/) (2024)
- [Assessing BERT's Syntactic Abilities](https://mlsanity.com/paper/1901.05287/) (2019)
- [Linear Representations of Sentiment in Large Language Models](https://mlsanity.com/paper/2310.15154/) (2023)
- [Causal Analysis of Syntactic Agreement Mechanisms in Neural Language Models](https://mlsanity.com/paper/2106.06087/) (2021)
- [Finding Universal Grammatical Relations in Multilingual BERT](https://mlsanity.com/paper/2005.04511/) (2020)
- [The emergence of number and syntax units in LSTM language models](https://mlsanity.com/paper/1903.07435/) (2019)
- [Interpreting Language Models with Contrastive Explanations](https://mlsanity.com/paper/2202.10419/) (2022)
- [Probing for the Usage of Grammatical Number](https://mlsanity.com/paper/2204.08831/) (2022)
- [CausalGym: Benchmarking causal interpretability methods on linguistic tasks](https://mlsanity.com/paper/2402.12560/) (2024)
- [Sparse Interventions in Language Models with Differentiable Masking](https://mlsanity.com/paper/2112.06837/) (2021)

## Built on by (most cited first, 24 of 24)

- [MultiBLiMP 1.0: A Massively Multilingual Benchmark of Linguistic Minimal Pairs](https://mlsanity.com/paper/2504.02768/) (2025)
- [Large Language Models Share Representations of Latent Grammatical Concepts Across Typologically Diverse Languages](https://mlsanity.com/paper/2501.06346/) (2025)
- [Paths Not Taken: Understanding and Mending the Multilingual Factual Recall Pipeline](https://mlsanity.com/paper/2505.20546/) (2025)
- [LLMs as a synthesis between symbolic and distributed approaches to language](https://mlsanity.com/paper/2502.11856/) (2025)
- [Different types of syntactic agreement recruit the same units within large language models](https://mlsanity.com/paper/2512.03676/) (2025)
- [Are Knowledge and Reference in Multilingual Language Models Cross-Lingually Consistent?](https://mlsanity.com/paper/2507.12838/) (2025)
- [Thank You, Stingray: Multilingual Large Language Models Can Not (Yet) Disambiguate Cross-Lingual Word Sense](https://mlsanity.com/paper/2410.21573/) (2024)
- [Discursive Circuits: How Do Language Models Understand Discourse Relations?](https://mlsanity.com/paper/2510.11210/) (2025)
- [Exploring the Translation Mechanism of Large Language Models](https://mlsanity.com/paper/2502.11806/) (2025)
- [Steering Language Models in Multi-Token Generation: A Case Study on Tense and Aspect](https://mlsanity.com/paper/2509.12065/) (2025)
- [How Instruction-Tuning Imparts Length Control: A Cross-Lingual Mechanistic Analysis](https://mlsanity.com/paper/2509.02075/) (2025)
- [Understanding or Memorizing? A Case Study of German Definite Articles in Language Models](https://mlsanity.com/paper/2601.09313/) (2026)
- [On the Emergence and Test-Time Use of Structural Information in Large Language Models](https://mlsanity.com/paper/2601.17869/) (2026)
- [The Grammar of Transformers: A Systematic Review of Interpretability Research on Syntactic Knowledge in Language Models](https://mlsanity.com/paper/2601.19926/) (2026)
- [Beyond Language: Format-Agnostic Reasoning Subspaces in Large Language Models](https://mlsanity.com/paper/2605.09496/) (2026)
- [Judge Circuits Explain Format-Induced Inconsistency in LLM-as-a-Judge](https://mlsanity.com/paper/2605.16023/) (2026)
- [Test, then Route: How Language Models Execute In-Context Conditional Rules Across Models and Languages](https://mlsanity.com/paper/2608.04183/) (2026)
- [Quantification and object perception in Multimodal Large Language Models and human linguistic cognition](https://mlsanity.com/paper/2511.08126/) (2025)
- [A Study of Crosslinguistic Influence in Language Models](https://mlsanity.com/paper/2601.21587/) (2026)
- [Exploring Language-Agnosticity in Function Vectors: A Case Study in Machine Translation](https://mlsanity.com/paper/2604.19678/) (2026)
- [Shared Circuits for Shared Grammar: Tracing Subject-Verb Agreement Across Languages](https://mlsanity.com/paper/2608.18545/) (2026)
- [Causal Interventions Reveal Typologically Organized Syntactic Mechanisms in Multilingual Language Models](https://mlsanity.com/paper/2608.28924/) (2026)
- [Shared circuits predict whether LLMs generalize across formats in arithmetic reasoning](https://mlsanity.com/paper/2609.04463/) (2026)
- [FTB Graph: Determining and Validating First-token Broadcasters and Language-Identity Head Circuits in Multilingual Language Models](https://mlsanity.com/paper/2609.30954/) (2026)

## Abstract (arXiv)

Several algorithms implemented by language models have recently been successfully reversed-engineered. However, these findings have been concentrated on specific tasks and models, leaving it unclear how universal circuits are across different settings. In this paper, we study the circuits implemented by Gemma 2B for solving the subject-verb agreement task across two different languages, English and Spanish. We discover that both circuits are highly consistent, being mainly driven by a particular attention head writing a `subject number' signal to the last residual stream, which is read by a small set of neurons in the final MLPs. Notably, this subject number signal is represented as a direction in the residual stream space, and is language-independent. We demonstrate that this direction has a causal effect on the model predictions, effectively flipping the Spanish predicted verb number by intervening with the direction found in English. Finally, we present evidence of similar behavior in other models within the Gemma 1 and Gemma 2 families.