High-dimensional data are extensively generated and utilized across various fields. Dimensionality reduction techniques, such as t-SNE, create low-dimensional embeddings that are easier to visualize. Recent research suggests that the dynamics of the embeddings during t-SNE optimization can reveal valuable information. Building on this insight, we developed visualizations that enable efficient visual analytics of t-SNE dynamics, helping users derive insights more effectively. Preliminary evaluations indicate that our visualizations not only make tasks easier to perform with greater confidence but also have the potential to uncover additional insights.