# Naturalness of Attention: Revisiting Attention in Code Language Models

arXiv: https://arxiv.org/abs/2311.13508 · HTML: https://mlsanity.com/paper/2311.13508/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 21 references that are in the map.
- First posted on arXiv 22 Nov 2023.

## Builds on (most cited first, 12 of 12)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Code Llama: Open Foundation Models for Code](https://mlsanity.com/paper/2308.12950/) (2023)
- [Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation](https://mlsanity.com/paper/1609.08144/) (2016)
- [CodeBERT: A Pre-Trained Model for Programming and Natural Languages](https://mlsanity.com/paper/2002.08155/) (2020)
- [CodeT5: Identifier-aware Unified Pre-trained Encoder-Decoder Models for Code Understanding and Generation](https://mlsanity.com/paper/2109.00859/) (2021)
- [CodeSearchNet Challenge: Evaluating the State of Semantic Code Search](https://mlsanity.com/paper/1909.09436/) (2019)
- [GraphCodeBERT: Pre-training Code Representations with Data Flow](https://mlsanity.com/paper/2009.08366/) (2020)
- [code2vec: Learning Distributed Representations of Code](https://mlsanity.com/paper/1803.09473/) (2018)
- [code2seq: Generating Sequences from Structured Representations of Code](https://mlsanity.com/paper/1808.01400/) (2018)
- [What Do They Capture? -- A Structural Analysis of Pre-Trained Language Models for Source Code](https://mlsanity.com/paper/2202.06840/) (2022)
- [An Exploratory Study on Code Attention in BERT](https://mlsanity.com/paper/2204.10200/) (2022)

## Built on by (most cited first, 1 of 1)

- [Neuron-Guided Interpretation of Code LLMs: Where, Why, and How?](https://mlsanity.com/paper/2512.19980/) (2025)

## Abstract (arXiv)

Language models for code such as CodeBERT offer the capability to learn advanced source code representation, but their opacity poses barriers to understanding of captured properties. Recent attention analysis studies provide initial interpretability insights by focusing solely on attention weights rather than considering the wider context modeling of Transformers. This study aims to shed some light on the previously ignored factors of the attention mechanism beyond the attention weights. We conduct an initial empirical study analyzing both attention distributions and transformed representations in CodeBERT. Across two programming languages, Java and Python, we find that the scaled transformation norms of the input better capture syntactic structure compared to attention weights alone. Our analysis reveals characterization of how CodeBERT embeds syntactic code properties. The findings demonstrate the importance of incorporating factors beyond just attention weights for rigorously understanding neural code models. This lays the groundwork for developing more interpretable models and effective uses of attention mechanisms in program analysis.