massOfai

Softmax Function

Neural Networks

Converts logits to probabilities

What is Softmax Function?

Outputs probability distribution over classes summing to 1. Used in multi-class classification.

Real-World Examples

  • •Final layer of image classifier
  • •LLM token prediction
  • •3-class output: [0.7, 0.2, 0.1]

When to Use This

Output layer for multi-class classification

Related Terms

Learn more about concepts related to Softmax Function

Classification

Categorizing data into predefined classes

Neural Network

Brain-inspired computing model with interconnected nodes

Probability

Measure of how likely events are to occur