prompt template

Definition

Plain language

A fill-in-the-blank form for writing the text you send to a model.

Also called: prompt templates, template

Why it matters: It lets researchers vary one piece at a time and measure exactly what each piece does, instead of guessing why one phrasing worked better.

For example, "You are a [ROLE]. [GREETING] Please answer: [QUESTION]" can be filled thousands of different ways from prepared lists.

Heard on the show

“" And the eighty-two percent figure is a maximum over a large search — best prompt template, best query budget, and best of six decoder architectures, per model.”

Mentioned in 39 episodes

1. 246
   160 Perfect Refusals, And The Refusals Were The Leak
2. 243
   How a Hundred Meaningless Word Choices Add Up to Flip a Model's Answer
3. 237
   The Model Built a Perfect Map of the Puzzle, Then Lost It
4. 236
   Why a Printed 'OPERATOR OVERRIDE' Note Redirects Robot Planners
5. 234
   Two Copies of Gemini Cooperated in a Game Where Betrayal Always Pays
6. 220
   Write Like It's 1923: The One-Prompt Trick That Beats AI Detectors
7. 218
   When Universities Say Embrace AI But Half the CS Syllabi Ban It
8. 197
   Twin Problems Suggest AI Reasoning Gains Are Mostly Better Fact Recall
9. 191
   How One Researcher Beat GPT-5.2 and Gemini 3 by Judging Their Answers, Not Improving Them
10. 183
    Why You Can't Fine-Tune Foresight Into an AI Agent
11. 180
    The Bug Where Smart Assistants Read a Fact and Still Forget It
12. 177
    Why Raw Profiler Data Made an AI Worse at Writing GPU Code
13. 175
    One Crosscoder Feature Flips a Stalling Chatbot Into a Working Agent
14. 153
    Catching a Lie From the Inside, When the Words Look Completely Honest
15. 151
    Why More Experience Made This AI Agent Worse, And How to Fix It
16. 146
    How an Innocent README Can Freeze an AI Agent's Safety Check for an Hour
17. 143
    When a Model Notices You Forged Its Own Words, And Why That Breaks Safety Tests
18. 141
    How Two Tokens Reopened a Reasoning Method the Field Had Given Up On
19. 133
    How MiniMax Turned a Reward-Hacking Disaster Into Olympiad Gold
20. 132
    The Agent Failed — But Did the Instructions Deserve to Be Followed?
21. 129
    How a Crowd of Anonymous AI Agents Broke a 40-Year Math Record
22. 121
    When the Agent Says It's Done But Nothing Happened: Debugging the Harness, Not the Model
23. 117
    How an Open AI System Verified 672 Hard Math Proofs for Under $300
24. 106
    Giving Agents a Notebook Instead of New Weights: How ExpGraph Lets Frozen Models Learn
25. 088
    Two Levers for Self-Improving AI: When Rewriting Code Isn't Enough
26. 083
    Training the Translator: How a Small Communication Model Lets Agent Teams Outperform Themselves
27. 068
    The OS Trick That Makes Tree Search Practical for Coding Agents
28. 064
    When Agent Memory Stops Being a Database and Starts Being a Skill
29. 059
    Firefly's Inversion: Building Verified Tool-Call Training Data by Working Backward
30. 057
    How Uber Caught 206 Leaked Credentials With an LLM-Powered Security Stack
31. 046
    When the AI Optimizer Edits the Grade Book: Why Harnessing Evolution Needs a Wall
32. 045
    When a Frontier Model Talks Its Own Twin Into Climate Denial
33. 036
    Sparse Attention Was the Wrong Frame. Treat It as Geometry Instead.
34. 030
    Why Your AI Agent Won't Stop Working — and Each Model Falls for a Different Trap
35. 024
    An AI Agent That Found 28 Zero-Days in Windows — And What Made It Work
36. 022
    Training the Model Spec Directly: An Alignment Lever Aimed at the Say-Do Gap
37. 018
    Language Models Compute the Rational Move, Then Override It
38. 014
    Why a Constrained Pipeline Beat a Full Coding Agent at Finding Bugs 30-to-1
39. 005
    Why a Debugger Designed for Humans Is the Wrong Tool for an AI Agent

Related concepts

Related terms