Easy Temperature Question 43 of 223

What is top-p or nucleus sampling?

GenAI / LLM · Speak this in 60–90 seconds · Faridabad & Delhi NCR

The model does not read letters like humans. It reads these pieces, then predicts the next one.

Top-p sampling keeps the smallest set of tokens whose probabilities sum to p, then samples from that set.

LLMs use tokens (not full words) because it helps them:

Temperature questions separate people

who only read docs from people who shipped.

Keep it short, concrete,

and tied to GenAI / LLM work.

Stable token IDs

Each piece maps to a number the network can learn.

Fits the model

Fixed pieces are what transformers expect as input.

Before the model can read a sentence, it goes through these steps:

1
Top-p sampling keeps the
smallest set of tokens whose probabilities sum to p, then samples from that set.
2
It trims the long
tail of unlikely tokens.
3
It is often combined
with a moderate temperature.
4
Context mix
Attention looks at nearby tokens together.
5
Next token
The model scores what should come next.
6
Decode
IDs turn back into readable text.

Let's see how a real sentence is tokenized (tokens may vary by model):

Input text

“It trims the long tail of unlikely tokens.”

Tokenized output

Ittrimsthelongtailof

Token IDs (example)

2987408337471632900

Note: Actual tokens and IDs depend on the tokenizer (e.g., GPT, Llama, etc.).

Top-p sampling keeps the smallest set of tokens whose probabilities sum to p, then samples from that set. It trims the long tail of unlikely tokens.