The Transformer architecture underpins modern large language models powering state-of-the-art text generation and AI applications. However, its complexity makes it difficult for non-experts to learn. Existing resources often lack interactivity, rely on static descriptions of simplified architectures, or fail to reflect models’ behavior with real data. To address this gap, we introduce Transformer Explainer, an interactive visualization tool for non-experts to learn Transformers. The tool integrates an overview illustrating the Transformer’s data flow with on-demand explanations that gradually reveal mathematical details. Smooth transitions across abstraction levels highlight the interplay between high-level structures and low-level operations. Running a live GPT-2 instance directly in the browser, Transformer Explainer empowers learners to experiment with custom input and hyperparameters without setup, observing next-token predictions in real time. A 90-participant user study showed that our tool offered significant advantages in improving user understanding and engagement. Transformer Explainer has attracted over 490,000 users.

Effective data visualizations enhance perception, support cognitive processing, and facilitate informed decision-making by aligning with human perceptual strengths. Conversely, poorly designed visualizations can impede comprehension, introduce interpretive bias, and diminish the perceived credibility of the conveyed message. This paper investigates the extent to which visual embellishments influence perceived message credibility in data visualizations. We conducted two crowdsourced experiments to examine both holistic and component-level effects of embellishment. In the first experiment, participants evaluated the relative credibility of plain bar charts versus two embellished variants-cartoon-style and image-style-across topics. Participants provided both comparative judgments and qualitative feedback. In the second experiment, we systematically isolated the influence of specific design elements-color, font, and bar style-on credibility perceptions through controlled variations. Our findings reveal that the impact of embellishments on perceived message credibility is complex and context-dependent. While certain embellishments, such as the use of color and image style bars, enhanced credibility, others-most notably hand-drawn fonts and cartoon-style bars-significantly undermined it. By operationalizing trust through the lens of message credibility, this work offers empirical insight into the design factors that shape viewers' perceptions. We conclude by proposing actionable design guidelines to support the creation of visualizations that are effective for communication and credible.

Transformers have revolutionized machine learning, yet their inner workings remain opaque to many. We present TRANSFORMER EXPLAINER, an interactive visualization tool designed for non-experts to learn about Transformers through the GPT-2 model. Our tool helps users understand complex Transformer concepts by integrating a model overview and smooth transitions across abstraction levels of math operations and model structures. It runs a live GPT-2 model locally in the user’s browser, empowering users to experiment with their own input and observe in real-time how the internal components and parameters of the Transformer work together to predict the next tokens. 125,000 users have used our open-source tool at https://poloclub.github.io/ transformer-explainer/.

As large language models (LLMs) see wider real-world use, understanding and mitigating their unsafe behaviors is critical.Interpretation techniques can reveal causes of unsafe outputs and guide safety, but such connections with safety are often overlooked in prior surveys.We present the first survey that bridges this gap, introducing a unified framework that connects safety-focused interpretation methods, the safety enhancements they inform, and the tools that operationalize them.Our novel taxonomy, organized by LLM workflow stages, summarizes nearly 70 works at their intersections.We conclude with open challenges and future directions.This timely survey helps researchers and practitioners navigate key advancements for safer, more interpretable LLMs.