How Programming Concepts and Neurons Are Shared in Code Language Models

• URL: http://arxiv.org/abs/2506.01074v1
• Date: Sun, 01 Jun 2025 16:24:13 GMT
• Title: How Programming Concepts and Neurons Are Shared in Code Language Models
• Authors: Amir Hossein Kargaran, Yihong Liu, François Yvon, Hinrich Schütze,
• Abstract summary: We perform a few-shot translation task on 21 PL pairs using two Llama-based models.<n>We observe that the concept space is closer to English (including PL keywords) and assigns high probabilities to English tokens in the second half of the intermediate layers.<n>We analyze neuron activations for 11 PLs and English, finding that while language-specific neurons are primarily concentrated in the bottom layers, those exclusive to each PL tend to appear in the top layers.
• Score: 55.22005737371843
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Several studies have explored the mechanisms of large language models (LLMs) in coding tasks, but most have focused on programming languages (PLs) in a monolingual setting. In this paper, we investigate the relationship between multiple PLs and English in the concept space of LLMs. We perform a few-shot translation task on 21 PL pairs using two Llama-based models. By decoding the embeddings of intermediate layers during this task, we observe that the concept space is closer to English (including PL keywords) and assigns high probabilities to English tokens in the second half of the intermediate layers. We analyze neuron activations for 11 PLs and English, finding that while language-specific neurons are primarily concentrated in the bottom layers, those exclusive to each PL tend to appear in the top layers. For PLs that are highly aligned with multiple other PLs, identifying language-specific neurons is not feasible. These PLs also tend to have a larger keyword set than other PLs and are closer to the model's concept space regardless of the input/output PL in the translation task. Our findings provide insights into how LLMs internally represent PLs, revealing structural patterns in the model's concept space. Code is available at https://github.com/cisnlp/code-specific-neurons.

Related papers

• The Emergence of Abstract Thought in Large Language Models Beyond Any Language [95.50197866832772]
  Large language models (LLMs) function effectively across a diverse range of languages.<n>Preliminary studies observe that the hidden activations of LLMs often resemble English, even when responding to non-English prompts.<n>Recent results show strong multilingual performance, even surpassing English performance on specific tasks in other languages.
  arXiv  Detail & Related papers  (2025-06-11T16:00:54Z)
• Converging to a Lingua Franca: Evolution of Linguistic Regions and Semantics Alignment in Multilingual Large Language Models [11.423589362950812]
  Large language models (LLMs) have demonstrated remarkable performance, particularly in multilingual contexts.<n>Recent studies suggest that LLMs can transfer skills learned in one language to others, but the internal mechanisms behind this ability remain unclear.<n>This paper provides insights into the internal workings of LLMs, offering a foundation for future improvements in their cross-lingual capabilities.
  arXiv  Detail & Related papers  (2024-10-15T15:49:15Z)
• Crosslingual Capabilities and Knowledge Barriers in Multilingual Large Language Models [62.91524967852552]
  Large language models (LLMs) are typically multilingual due to pretraining on diverse multilingual corpora.<n>But can these models relate corresponding concepts across languages, i.e., be crosslingual?<n>This study evaluates state-of-the-art LLMs on inherently crosslingual tasks.
  arXiv  Detail & Related papers  (2024-06-23T15:15:17Z)
• How do Large Language Models Handle Multilingualism? [81.15060972112563]
  This study explores how large language models (LLMs) handle multilingualism. LLMs initially understand the query, converting multilingual inputs into English for task-solving. In the intermediate layers, they employ English for thinking and incorporate multilingual knowledge with self-attention and feed-forward structures.
  arXiv  Detail & Related papers  (2024-02-29T02:55:26Z)
• Language-Specific Neurons: The Key to Multilingual Capabilities in Large Language Models [117.20416338476856]
  Large language models (LLMs) demonstrate remarkable multilingual capabilities without being pre-trained on specially curated multilingual parallel corpora. We propose a novel detection method, language activation probability entropy (LAPE), to identify language-specific neurons within LLMs. Our findings indicate that LLMs' proficiency in processing a particular language is predominantly due to a small subset of neurons.
  arXiv  Detail & Related papers  (2024-02-26T09:36:05Z)
• Joint Prompt Optimization of Stacked LLMs using Variational Inference [66.04409787899583]
  Large language models (LLMs) can be seen as atomic units of computation mapping sequences to a distribution over sequences. By stacking two such layers and feeding the output of one layer to the next, we obtain a Deep Language Network (DLN) We show that DLN-2 can reach higher performance than a single layer, showing promise that we might reach comparable performance to GPT-4.
  arXiv  Detail & Related papers  (2023-06-21T18:45:56Z)
• Efficiently Aligned Cross-Lingual Transfer Learning for Conversational Tasks using Prompt-Tuning [98.60739735409243]
  Cross-lingual transfer of language models trained on high-resource languages like English has been widely studied for many NLP tasks. We introduce XSGD for cross-lingual alignment pretraining, a parallel and large-scale multilingual conversation dataset. To facilitate aligned cross-lingual representations, we develop an efficient prompt-tuning-based method for learning alignment prompts.
  arXiv  Detail & Related papers  (2023-04-03T18:46:01Z)
• ERNIE-Code: Beyond English-Centric Cross-lingual Pretraining for Programming Languages [37.60016772021422]
  Software engineers working with the same programming language (PL) may speak different natural languages (NLs) and vice versa. Recent studies have demonstrated the effectiveness of generative pre-training in computer programs, yet they are always English-centric. We release ERNIE-Code, a unified pre-trained language model for 116 NLs and 6 PLs.
  arXiv  Detail & Related papers  (2022-12-13T17:21:44Z)
• Programming Language Agnostic Mining of Code and Language Pairs with Sequence Labeling Based Question Answering [15.733292367610627]
  Mining aligned natural language (NL) and programming language (PL) pairs is a critical task to NL-PL understanding. We propose a Sequence Labeling based Question Answering (SLQA) method to mine NL-PL pairs in a PL-agnostic manner.
  arXiv  Detail & Related papers  (2022-03-21T05:33:59Z)
• Exploiting Language Relatedness for Low Web-Resource Language Model Adaptation: An Indic Languages Study [14.34516262614775]
  We argue that relatedness among languages in a language family may be exploited to overcome some of the corpora limitations of LRLs. We focus on Indian languages, and exploit relatedness along two dimensions: (1) script (since many Indic scripts originated from the Brahmic script) and (2) sentence structure.
  arXiv  Detail & Related papers  (2021-06-07T20:43:02Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.