Skip to content
AImpact
IT EN
Inference Intermediate Also known as: Nucleus Sampling · Campionamento a nucleo

Top-p Sampling

/top-pee sampling/

A strategy that picks the next token from the smallest set of candidates whose cumulative probability exceeds a threshold p (e.g. 0.9).

ShareLinkedInX

In practice

It adapts the candidate set to context: few options when the model is confident, many when it is unsure. It is the most-used parameter in APIs (`top_p` on OpenAI, Anthropic, etc.) to tune creativity without losing coherence. Typical values sit between 0.8 and 0.95.

Related terms

Top-k Sampling Temperature Logits Softmax

Seen in the wild

0 entries mentioning it

No archive entry mentions it explicitly. Appears in broader contexts.

← All terms