Parameter
An adjustable number that determines how a model or a machine behaves. In AI it means the strengths of the connections between neurons, the weights, which are set by training.
In a neural network every connection between neurons carries a number called a weight. Training is the process of nudging those numbers, little by little, while showing the network a great many examples[1]. A large language model has hundreds of billions to trillions of parameters, and everything it knows and can do is held in that array of numbers.
More broadly, the settings that determine how a machine or an experiment runs are also called parameters. A quantum computer, for instance, has a large number of control parameters, such as the strength and timing of the signals that operate its qubits, which must be continually tuned to stop them drifting.
Related terms
- Neural networkA system of simple computing units called neurons, connected in many layers, that learns from data how strong each connection should be and so becomes able to make complex judgements.
- NeuronThe smallest unit of computation in a neural network. It weights and sums its inputs and produces an output depending on the result.
- Large language modelAn AI trained on vast amounts of text that generates language by predicting the next word. The core of ChatGPT, Claude and similar systems.