In this project, we aim to investigate quantum neural networks from the perspective of how quantum information is distributed throughout the circuit. In particular, we will analyze several information-theoretic quantities such as mutual information, tripartite information, and entanglement entropy and examine how these measures evolve during the training of a quantum ansatz for different computational tasks. We will begin by implementing these analyses for discrete-variable quantum neural networks. Subsequently, we will extend the framework to more general settings, including continuous-variable quantum neural networks, in order to explore how information dynamics differ across architectures and representations.