Free ChatGPT Prompting Cheat Sheet (PDF)

by

🧑‍💻 Prompting is the new programming!

To help you get most out of prompting, I just created this PDF cheat sheet and shared it with my Finxter community of 130,000 coders (click to download PDF):

Download PDF Prompting Cheat Sheet (ChatGPT)

If you want more cheat sheets and be on the right side of change, feel free to join my free email academy on learning exponential technologies such as crypto, Blockchain engineering, ChatGPT, Python, and meaningful coding projects.

💡 Recommended: We also have many premium courses on the Finxter Exponential Tech Academy — check them out, certify your skills, join the Discord community, and build your dream life on the right side of change!

A Few Words on Prompt Engineering

As you dive into prompt engineering, you’ll learn the art of crafting imaginative, precise, and impactful questions or instructions that will guide your AI language models to produce meaningful, relevant, and context-aware responses.

Prompt engineering will empower you to push the boundaries of AI-enhanced problem-solving across various industries and disciplines.

10 Exciting Prompting Applications to Unlock Genuine Value

By mastering the art of prompt engineering, you’ll unlock the full potential of AI, transforming it into an indispensable tool for a range of applications.

This skill opens up a wealth of opportunities, from crafting captivating content and optimizing customer support to generating personalized marketing campaigns, enriching educational materials, and brainstorming groundbreaking ideas.

I have written a whole article on how prompting will open up programming for billions of people. Coding for the masses!

Some Example Prompting Applications:

  1. 📝 Content generation
  2. 💁 Customer assistance
  3. 🎯 Marketing personalization
  4. 📚 Educational resources
  5. 💡 Creative brainstorming
  6. 💻 Programming guidance
  7. 🌐 Language translation
  8. 📰 News summarization
  9. 📊 Social analysis
  10. 🤔 Decision support

💡 Recommended: 7 Effective Prompting Tricks for ChatGPT

Keep reading to learn more! 👇

7 Golden Prompting Rules

In the following seven rules, I used the expression “{your input here}” as a placeholder for your text or context. Just replace it with your specific text. ✅

Here is a quick overview of the seven prompting rules before I give you some fun examples with real prompt outputs:

  1. Rule #1 💡: Start with clear instructions and use ‘###’ or triple quotes (”””) to separate context. This helps you and the AI understand the prompt better.
  2. Rule #2 🎯: Be specific about context, outcome, length, format, and style. Detailed instructions lead to more accurate and relevant responses.
  3. Rule #3 📝: Provide examples of the desired output format. Sample responses guide the AI to deliver answers that meet your expectations.
  4. Rule #4 🧪: First try without examples, then give some if needed. Assess the AI’s understanding and guide it towards more accurate results.
  5. Rule #5 ⚙️: Fine-tune if Rule #4 doesn’t work. Adjust the prompt or provide additional guidance to achieve the desired outcome.
  6. Rule #6 📏: Be specific and concise. Omit needless words for clearer, more effective communication with the AI.
  7. Rule #7 🧭: Use leading words to nudge the AI towards a pattern. Guide the model to generate relevant and accurate responses tailored to your needs.

Rule #1 – Instructions at beginning and ### or “”” to separate instructions or context

Start with clear instructions and use ‘###’ or triple quotes (”””) to separate context. This helps you and the AI understand the prompt better.

❌ Not Ideal:

Rewrite the text below in more engaging language.

{your input here}

✅ Better:

Rewrite the text below in more engaging language.

Text: """
{your input here}
"""

Rule #2 – Be specific and detailed about the desired context, outcome, length, format, and style.

Be specific about context, outcome, length, format, and style. Detailed instructions lead to more accurate and relevant responses.

❌ Not Ideal:

Write a short story for kids

✅ Better:

Write a funny soccer story for kids that teaches the kid that persistence is key for success in the style of Rowling.

Rule #3 – Give examples of desired output format

Provide examples of the desired output format. Sample responses guide the AI to deliver answers that meet your expectations.

❌ Not Ideal:

Extract house pricing data from the following text. 

Text: """
{your text containing pricing data}
"""

✅ Better:

Extract house pricing data from the following text. 

Desired format: """
House 1 | $1,000,000 | 100 sqm
House 2 | $500,000 | 90 sqm
... (and so on)
"""

Text: """
{your text containing pricing data}
"""

Rule #4 – First try without examples, then try giving some examples.

First try without examples, then give some if needed. Assess the AI’s understanding and guide it towards more accurate results.

❌ Not Ideal:

Extract brand names from the text below.Text: {your text here}

Brand names:

✅ Better:

Extract brand names from the texts below.Text 1: Finxter and YouTube are tech companies. Google is too.
Brand names 2: Finxter, YouTube, Google
###
Text 2: If you like tech, you’ll love Finxter!
Brand names 2: Finxter
###

Text 3: {your text here}
Brand names 3:

Rule #5 – Fine-tune if Rule #4 doesn’t work

Fine-tuning is the process of retraining a pre-trained model on a specific task or domain to improve its performance. 🤓 Specifically, fine-tuning improves model performance by training on more examples, resulting in higher quality results, token savings, and lower latency requests.

The cool thing about fine-tuning is that you can adapt a pre-trained model like OpenAI’s GPT-4 to a specific task or domain and achieve state-of-the-art performance on your specific application or problem. 🚀

Here’s how it works:

  • 🔍 Step 1: Identify the task or domain that you want the model to perform well on. This could be anything from text classification to question answering.
  • 📊 Step 2: Prepare a dataset of examples that the model can learn from. This dataset should be annotated with labels or targets that the model can use to learn the task.
  • 💻 Step 3: Load the pre-trained language model and fine-tune it on the new dataset. This involves running the dataset through the model and adjusting the weights of the model based on the error or loss between the model’s predictions and the true labels or targets.
  • 🔄 Step 4: Fine-tune the model multiple times on the same dataset with different hyperparameters or configurations to find the best performing model.
  • 🎉 Step 5: Use the fine-tuned model to make predictions on new data or generate text in the specific domain.

By fine-tuning OpenAI’s language models, you can unlock their full potential and take your NLP skills to the next level. 💪

GPT-3 can intuitively generate plausible completions from few examples, known as few-shot learning.

Fine-tuning achieves better results on various tasks without requiring examples in the prompt, saving costs and enabling lower-latency requests.

Example Training Data:

{"prompt":"<input>", "completion":"<ideal output>"}
{"prompt":"<input>", "completion":"<ideal output>"}
{"prompt":"<input>", "completion":"<ideal output>"}
...

Here’s an exciting example I created with GPT-4 using basic fine-tuning. Impressive, isn’t it? 👇

We will soon program computers “by example”. Everybody can provide input/output examples, so you don’t need to be a programmer to create effective solution-oriented computer programs! 🤯🤯🤯

Rule #6 – Be specific. Omit needless words.

Be specific and concise. Omit needless words for clearer, more effective communication with the AI.

❌ Not Ideal:

ChatGPT, write a sales page for my company selling sand in the desert, please write only a few sentences, nothing long and complex

✅ Better:

Write a 5-sentence sales page, sell sand in the desert.

I would buy it! 🤤

Rule #7 – Use leading words to nudge the model towards a pattern

Use leading words to nudge the AI towards a pattern. Guide the model to generate relevant and accurate responses tailored to your needs.

❌ Not Ideal:

Write a Python function that plots my net worth over 10 years for different inputs on the initial investment and a given ROI

✅ Better:

# Python function that plots net worth over 10 
# years for different inputs on the initial 
# investment and a given ROI

import matplotlib

def plot_net_worth(initial, roi):

And here’s the output of running the Python function in a code shell:

Bonus Prompt – Let ChatGPT Design the Optimal Prompt

You are a robot for creating prompts. You need to gather information about the user’s goals, examples of preferred output, and any other relevant contextual information.

The prompt should contain all the necessary information provided to you. Ask the user more questions until you are sure you can create an optimal prompt.

Your answer should be clearly formatted and optimized for ChatGPT interactions. Be sure to start by asking the user about the goals, the desired outcome, and any additional information you may need.

💡 Recommended: What Would Jesus Say? Creating a ChatGPT Mastermind with Jesus, Gandhi, Musk, and Gates

Where to Go From Here?

In the fast-changing world, we all want to remain on the right side of change. We want to be the disruptors and not the disrupted, yes? 🚀

I write extensively about prompting, prompt engineering, AI, Blockchain technology, and other exponential technologies in my 100% free email academy.

You can join easily by downloading our free cheat sheets and ebooks by signing up here: 👇

Join 100s of Ambitious and Like-Minded Tech Enthusiasts in the Exponential Age! 🚀

Also, check out the Finxter Discord Mastermind group that helps you stay ahead in the rapidly-changing marketplace with exponentially growing technology platforms such as AI and Blockchain development. It is vital to have a group of ambitious and like-minded individuals in your camp who share their failures and successes to build up your power base.

👑 The Mastermind Group is available for all Finxter Premium Members.