AI Blog

AI Blog

by Michele Laurelli

Back to Glossary

Parameter

Concept
Definition

Learnable values in a model that are optimized during training (weights and biases).

Parameters define the model. Number of parameters indicates model capacity. Large models have billions of parameters.

Examples

1

GPT-3: 175B parameters

2

Weight and bias parameters

3

Model size

Related Terms

Weight

Learnable parameters in neural networks that determine the strength of connections between neurons.

Bias Term

An additional learnable parameter in neural networks that allows shifting the activation function.

Michele Laurelli - AI Research & Engineering