Softmax is a mathematical function that converts a vector of raw scores (logits) into a probability distribution, where each value is between 0 and 1 and all values sum to 1. It is the standard final layer in classification neural networks, turning the model's raw outputs into interpretable class probabilities. For example, a model predicting whether an email is spam might output logits [2.1, 0.3, -1.5], which softmax converts to probabilities [0.88, 0.11, 0.01].