# A Multiscale Visualization of Attention in the Transformer Model

arXiv: https://arxiv.org/abs/1906.05714 · HTML: https://mlsanity.com/paper/1906.05714/

## Key facts

- Cited by 292 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 12 Jun 2019.

## Built on by (most cited first, 30 of 292)

- [GPT Understands, Too](https://mlsanity.com/paper/2103.10385/) (2021)
- [SpQR: A Sparse-Quantized Representation for Near-Lossless LLM Weight Compression](https://mlsanity.com/paper/2306.03078/) (2023)
- [Pre-Trained Language Models for Interactive Decision-Making](https://mlsanity.com/paper/2202.01771/) (2022)
- [Synthesis of Compositional Animations from Textual Descriptions](https://mlsanity.com/paper/2103.14675/) (2021)
- [How do Large Language Models Handle Multilingualism?](https://mlsanity.com/paper/2402.18815/) (2024)
- [Understanding and Overcoming the Challenges of Efficient Transformer Quantization](https://mlsanity.com/paper/2109.12948/) (2021)
- [Toward Transparent AI: A Survey on Interpreting the Inner Structures of Deep Neural Networks](https://mlsanity.com/paper/2207.13243/) (2022)
- [Why do LLMs attend to the first token?](https://mlsanity.com/paper/2504.02732/) (2025)
- [SparQ Attention: Bandwidth-Efficient LLM Inference](https://mlsanity.com/paper/2312.04985/) (2023)
- [HealthGPT: A Medical Large Vision-Language Model for Unifying Comprehension and Generation via Heterogeneous Knowledge Adaptation](https://mlsanity.com/paper/2502.09838/) (2025)
- [Circuit Component Reuse Across Tasks in Transformer Language Models](https://mlsanity.com/paper/2310.08744/) (2023)
- [A Survey on Gender Bias in Natural Language Processing](https://mlsanity.com/paper/2112.14168/) (2021)
- [HiT: Hierarchical Transformer with Momentum Contrast for Video-Text Retrieval](https://mlsanity.com/paper/2103.15049/) (2021)
- [BadPre: Task-agnostic Backdoor Attacks to Pre-trained NLP Foundation Models](https://mlsanity.com/paper/2110.02467/) (2021)
- [Trustworthiness in Retrieval-Augmented Generation Systems: A Survey](https://mlsanity.com/paper/2409.10102/) (2024)
- [Zero Shot Health Trajectory Prediction Using Transformer](https://mlsanity.com/paper/2407.21124/) (2024)
- [Copy Suppression: Comprehensively Understanding an Attention Head](https://mlsanity.com/paper/2310.04625/) (2023)
- [Iteratively Prompt Pre-trained Language Models for Chain of Thought](https://mlsanity.com/paper/2203.08383/) (2022)
- [Unveiling and Harnessing Hidden Attention Sinks: Enhancing Large Language Models without Training through Attention Calibration](https://mlsanity.com/paper/2406.15765/) (2024)
- [BiBERT: Accurate Fully Binarized BERT](https://mlsanity.com/paper/2203.06390/) (2022)
- [On the Role of Attention Heads in Large Language Model Safety](https://mlsanity.com/paper/2410.13708/) (2024)
- [Interactive and Visual Prompt Engineering for Ad-hoc Task Adaptation with Large Language Models](https://mlsanity.com/paper/2208.07852/) (2022)
- [FormNet: Structural Encoding beyond Sequential Modeling in Form Document Information Extraction](https://mlsanity.com/paper/2203.08411/) (2022)
- [The Road to Know-Where: An Object-and-Room Informed Sequential BERT for Indoor Vision-Language Navigation](https://mlsanity.com/paper/2104.04167/) (2021)
- [Neural Media Bias Detection Using Distant Supervision With BABE -- Bias Annotations By Experts](https://mlsanity.com/paper/2209.14557/) (2022)
- [Why So Gullible? Enhancing the Robustness of Retrieval-Augmented Models against Counterfactual Noise](https://mlsanity.com/paper/2305.01579/) (2023)
- [Regression Transformer: Concurrent sequence regression and generation for molecular language modeling](https://mlsanity.com/paper/2202.01338/) (2022)
- [AttentionViz: A Global View of Transformer Attention](https://mlsanity.com/paper/2305.03210/) (2023)
- [polyBERT: A chemical language model to enable fully machine-driven ultrafast polymer informatics](https://mlsanity.com/paper/2209.14803/) (2022)
- [Investigating Explainability of Generative AI for Code through Scenario-based Design](https://mlsanity.com/paper/2202.04903/) (2022)

## Abstract (arXiv)

The Transformer is a sequence model that forgoes traditional recurrent architectures in favor of a fully attention-based approach. Besides improving performance, an advantage of using attention is that it can also help to interpret a model by showing how the model assigns weight to different input elements. However, the multi-layer, multi-head attention mechanism in the Transformer model can be difficult to decipher. To make the model more accessible, we introduce an open-source tool that visualizes attention at multiple scales, each of which provides a unique perspective on the attention mechanism. We demonstrate the tool on BERT and OpenAI GPT-2 and present three example use cases: detecting model bias, locating relevant attention heads, and linking neurons to model behavior.