Sulba
ReviewSign inAccount
Theme
Editor theme
Light
Dark

Beginner · Text as numbers

Logits and sampling

Ask a small language model whether products match shoppers' searches, and read its answers from the logits: turn them into probabilities, reshape them with temperature, top-k and top-p, draw answers as generation does, and ship probabilities that mean what they say.

You build
judge.py, then sampler.py
Stages
3 Build, 1 Catch, 1 Ship
Time
About 3.5 hours

Loading the workspace…