But What Are Transformers?

A comprehensive walkthrough of how Transformer neural networks work, covering tokenization, token embeddings, the attention mechanism (including queries, keys, and values), multi-head attention, positional encoding, residual connections, layer normalization, and the encoder-decoder architecture. Also compares encoder-only (BERT), decoder-only (GPT, LLaMA), and encoder-decoder (T5, BART) model families and their respective use cases.