To enhance explanations about artificial intelligence systems, visualizations can be employed for the task of explaining deep learning models. Since social explanations are incremental and multimodal (Chap. 1 ), visualizations serve as both an incremental component and an additional modality, such as alongside textual or verbal explanations. Thereby, visualizations can help create more intuitive explanations that are adapted to the needs of an explainee. In this chapter, we focus on visualizing local post hoc explanations that can explain specific decisions made by a deep-learning-based system. These explanations can, for example, answer questions on why a decision that affects the user was made. We review how these visualizations are currently produced and how they can be used in social explainable artificial intelligence. To evaluate visualizations in explanations for black-box deep learning systems, we suggest to focus on the social aspects of the explanation process.