probability

A number between 0 and 1 measuring how likely a token or sequence is, according to the model.

A number $p \in [0,1]$ assigned by a model to an event -- most often, to the next token given everything generated so far. A language model's whole job is producing a probability for every token in its vocabulary at each step.

Examples

A model might assign probability 0.4 to "cat" and 0.01 to "airplane" after the words "The pet was a".