Model Parameters
Model parameters are the internal numerical values, mainly weights and biases, that a machine learning model learns from data and uses to make predictions.
What is Model Parameters?
The number of parameters is a rough measure of a model's size and capacity; large language models have from a few billion to over a trillion parameters. Parameters are learned automatically, unlike hyperparameters, which are chosen by developers.
How it works
Training starts with random values and repeatedly nudges each parameter in the direction that reduces the loss on the training data.
← All Artificial Intelligence & Superintelligence definitions · ← Back to the A–Z definitions