Temperature and sampling
Sampling parameters decide how the next token is picked. Low temperature for extraction and classification, higher for anything that should feel varied.
- 8m
- Intermediate
Video by DataMListic · YouTube
Check your understanding
3 questions. Unlocks when you finish the video.
Overview
What this lesson covers
- Trade determinism against variety with temperature
- Know what top-p changes and when to touch it
- Pick settings per task rather than globally
This lesson sits in LLM Fundamentals, part of Building AI Apps with LLMs. It assumes what came before it and leads directly into the next lesson in the module.
In this lesson you will:
- Trade determinism against variety with temperature
- Know what top-p changes and when to touch it
- Pick settings per task rather than globally
Resources
Your notes will live here
Note taking is not available yet. Nothing you type would be saved, so the tab stays read-only for now.