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…
Loading the workspace…