Sampling
Also: decoding, decoding method, decoding strategy, probability distribution
How the model picks the next word from its list of likely candidates. It can always take the top one, or roll the dice among the top few.
Always picks the most likely next word. Ask twice, get the same answer twice.
Sometimes picks a less likely word. Ask twice, get two different answers.
Neither setting makes the answer more true. Low temperature makes mistakes repeatable; high temperature makes them varied.
The fuller definition
Sampling selects an output from a probability distribution. In language generation, a decoding strategy selects successive tokens from model scores.
Go deeper
Invented probabilities. Always picking the top candidate gives the same wording every run; sampling sometimes picks a lower one. Neither setting checks whether the sentence is true.
A concrete example
Different runs may choose different plausible next words for the same sentence.
The distinction that matters
Variation can be useful for drafting but does not itself provide independent evidence.
Related terms
Temperature
A setting that affects how varied the output may be. Lower values often make it less varied. They do not ensure the same answer each time.
Training & generationTop-p sampling
A setting that narrows the next-token choices. It keeps a group with a set share of the total chance. Lower values usually narrow that group.
FoundationsToken
A piece of text, often part of a word, that the model actually works with. Models read and write tokens, not whole words.
Where it is taught
How models learn from examplesSource and further reading
Hugging Face · LLM courseThe definition and example are written for this library. The link gives background, official documentation or legal authority as labelled. Legal scope depends on the jurisdiction and current text; product features depend on release and settings.