AIExplainer
Machine Learning Mathematics Intermediate 1 min read

What is logits?

The raw, unnormalized scores produced by a model before applying an activation function

Logits are the output values from a neural network before they are converted into probabilities. They represent the relative likelihood of each possible outcome, but are not yet probabilities themselves.

Think of logits like unadjusted exam scores. Just as exam scores need to be scaled and normalized to become percentages, logits need to be passed through an activation function to become probabilities.

In image classification, logits might represent the relative likelihood of an image being a cat, dog, or bird. The softmax function would then be applied to these logits to produce a probability distribution, such as 70% cat, 20% dog, and 10% bird.

Logits are used as the input to an activation function, such as the softmax function, to produce a probability distribution over all possible outcomes.

A common misconception is that logits are the same as probabilities. However, logits are not normalized and do not add up to 1, whereas probabilities do.

The term 'logit' was first used in statistics in the 1940s, but its use in machine learning and deep learning has become more widespread in recent years.

raw scores unnormalized scores pre-activation values

Three products for different needs — explore what’s relevant to you.