n the upcoming year, we plan to further evaluate and analyze the training process of Spiking Neural Networks (SNNs). SNNs are governed by temporal dynamics and can be viewed as dynamical systems, which makes the properties of their training process particularly important in terms of stability, convergence, and robustness under different conditions. The main focus will be placed on the evaluation of parameterization principles and on understanding how different configurations influence learning dynamics over time. In particular, we aim to systematically study how network size, depth, and parameter choices affect the training process with an emphasis on how temporal components contribute to overall behavior and performance. The computational effort will be directed toward training and evaluating a large number of network configurations under varying settings. We will consider a range of architectures and network sizes in order to better understand how scaling influences training behavior, as well as how different temporal processing choices affect learning outcomes. To achieve this, We plan to conduct a large number of parallel training runs for data collection and analysis. This large-scale experimental setup is necessary to cover the breadth of configurations required for a systematic study and goes beyond what can be efficiently handled with local computational resources.