This paper introduces the Token Space framework, a novel mathematical construct designed to enhance the interpretability and effectiveness of deep learning models through the application of category theory. By establishing a categorical structure at the Token level, we provide a new lens through which AI computations can be understood, emphasizing the relationships between tokens, such as grouping, order, and parameter types. We explore the foundational methodologies of the Token Space, detailing its construction, the role of construction operators and initial categories, and its application in analyzing deep learning models, specifically focusing on attention mechanisms and Transformer architectures. The integration of category theory into AI research offers a unified framework to describe and analyze computational structures, enabling new research paths and development possibilities. Our investigation reveals that the Token Space framework not only facilitates a deeper theoretical understanding of deep learning models but also opens avenues for the design of more efficient, interpretable, and innovative models, illustrating the significant role of category theory in advancing computational models.

Deep convolutional neural networks (CNNs) have demonstrated impressive performance on visual object classification tasks. In addition, it is a useful model for predication of neuronal responses recorded in visual system. However, there is still no clear understanding of what CNNs learn in terms of visual neuronal circuits. Visualizing CNN's features to obtain possible connections to neuronscience underpinnings is not easy due to highly complex circuits from the retina to higher visual cortex. Here we address this issue by focusing on single retinal ganglion cells with a simple model and electrophysiological recordings from salamanders. By training CNNs with white noise images to predicate neural responses, we found that convolutional filters learned in the end are resembling to biological components of the retinal circuit. Features represented by these filters tile the space of conventional receptive field of retinal ganglion cells. These results suggest that CNN could be used to reveal structure components of neuronal circuits.