What Is a Transformer? A Visual Overview for Beginners

What is a Transformer, and how does it turn one sentence into another? This beginner-friendly overview follows the English sentence "The cat sleeps" as it is translated into French. You will learn what a neural network and its learned weights are, how text is split into tokens, how an embedding turns tokens into vectors, why position information is added, and how attention lets words gather context, with small worked calculations. Then see the full original encoder-decoder Transformer: what self-attention, feed-forward networks, Add & Norm (residual connection and layer normalization), cross-attention and softmax each do, how blocks are stacked, how the decoder writes one token at a time, and how training improves the weights. No prior knowledge needed. Each part gets its own deeper video in this series.