I have a bunch of Jupyter notebooks that I made to satisfy my own curiosity. You can look at them here

• Gradient Visualization This is based on my curiosity of how the weights of a neural network change during the training process. Here I train a very simple neural net on MNIST, and I try to do some analysis on how the weights are changing. I do this by creating a simple dimensionality reduction method.
• Diffusion I tried to create a very simple diffusion model