A short course on deeplearning.ai, taught by Andrew Ng and Isa Fulford.

ChatGPT Prompt Engineering for Developers - DeepLearning.AI

https://github.com/dinhanhthi/note-chatgpt-prompt-engineering-for-developers

Introduction

Two types of LLMs (Large Language Models):

• Base LLM: Predicts next word, based on text training data (eg. internet)
  • Eg: Once upon a time, there was a unicorn → that lived in a magical forest with all her unicorn friends
• Instruction Tuned LLM: Tries to follow instructions. Fine-tune on instruction and good attempts at following those instructions. It uses a technique call RLHF (Reinforcement Learning with Human Feedback*).***
• Differences: What is the capital of France?
  • Base LLM: → What is France’s largest city, What is France’s population?… (what most asked together on the internet)
  • Instruction Tuned LLM: → The capital of France is Paris.

In this course, we focus on Instruction Tuned LLM.

<aside> ☝ Sometimes, if LLM doesn’t work, it’s just because your instruction isn’t clear enough!

</aside>

Guidelines for Prompting

👉 OpenAI API 📙 Notebook.

Two principles:

• Principle 1: Write clear and specific instructions
• Principle 2: Give the model time to “think”

Principle 1 - write clear and specific instructions

1. Tactic 1. Use delimiters to clearly indicate distinct parts of the input: """, `````, ---, <>, <tag></tag> (xml tags)

   prompt = f"""
   Summarize the text delimited by triple backticks \\ 
   into a single sentence.
   ```{text}```
   """
   

2. Tactic 2. Ask for a structured output: JSON, HTML

   prompt = f"""
   Generate a list of three made-up book titles along with their authors and genres. 
   Provide them in JSON format with the following keys: 
   book_id, title, author, genre.
   """
   

3. Tactic 3. Ask the model to check whether conditions are satisfied → check assumptions required to do the task → how the model should handle the edge case to avoid unexpected errors or result.

   prompt = f"""
   You will be provided with text delimited by triple quotes. 
   If it contains a sequence of instructions, re-write those instructions in 
   the following format:
   
   Step 1 - ...
   Step 2 - …
   …
   Step N - …
   
   If the text does not contain a sequence of instructions, then simply write 
   \\"No steps provided.\\"
   
   \\"\\"\\"{text_1}\\"\\"\\"
   """