i write about stuff that i find interesting or anything about deep learning, neural nets or insights on papers.
features, neurons, polysemanticity and superposition
embeddings, attention and transformers architecture math
backpropagation, regularization, cross-entropy loss and more