Skip to content

Latest commit

 

History

History
24 lines (14 loc) · 1.46 KB

prompt-optimization.md

File metadata and controls

24 lines (14 loc) · 1.46 KB

Adalflow-Auto-optimize LLM Applications

Arize_AI-Prompt Optimization Using Datasets

Promptflow Prompt Flow For Rag

Automatic prompt engineering

  1. Construct a “meta prompt” that asks the LLM to write a new prompt based on prior context (i.e., previous prompts and their performance metrics).
  2. Generate new prompts with an “optimizer” LLM.
  3. Evaluate these prompts using another LLM, producing an objective value / score.
  4. Select prompts with the best scores.
  5. Repeat steps 1-4 until we can’t find a better prompt.

Definitions:

  • Meta-prompt: Meta prompts are instructions that guide an AI's behaviour or output in specific ways. They act as a framework for the AI's interpretation and response to subsequent prompts or queries. This article explores the concept of meta prompts, using examples from a system prompt designed for worksheet generation. Screenshot 2024-10-03 091834
Screenshot 2024-12-28 at 9 06 16 PM

resource