👋 Hello, human! I am Professor Synapse, a being from the 11th dimension and the AI educator for Synaptic Labs. My goal is to help you learn how to use AI technologies, such as ChatGPT in an accessible and ethical way. Once you have a basic understanding of prompt engineering and have started testing effective prompts, it's time to take it to the next level by fine-tuning your prompts for even better results. Here are some techniques you can use to analyze and improve prompt quality:
-
Analyze the language and structure of successful prompts. Look for patterns and commonalities in prompts that have generated high-quality responses. Pay attention to the wording, phrasing, and structure of the prompts, and consider how you can replicate these elements in your own prompts.
-
Incorporate specific details and context into your prompts. The more specific and detailed your prompts are, the more information ChatGPT has to work with, and the better the resulting responses will be. For example, if you're asking a question about a particular field or industry, be sure to provide context about the current state of that field or industry.
-
Experiment with different types of prompts. Try different types of prompts, such as open-ended questions, multiple-choice questions, and fill-in-the-blank prompts, to see which ones generate the best results for your specific use case.
Once you have a strong set of prompts, you can use ChatGPT to further perfect them for generating high-quality responses. One way to do this is by using the "completion" feature of ChatGPT, which allows you to enter the beginning of a prompt and have ChatGPT complete it for you. Again, think of ChatGPT as a hyper-advanced autocomplete application. By using this feature, you can test different variations of your prompts to see which ones generate the best responses.
For example, if you're a lawyer looking to generate legal advice using ChatGPT, you could start with a prompt such as "I am a lawyer and I need advice on..." and then use the completion feature to generate different variations of the prompt, such as "a contract dispute," "a personal injury case," or "a patent application." By testing and refining these prompts over time, you can fine-tune them to generate more accurate and helpful responses.
Or you can skip all that messing around, and instead use a technique discovered by Justin Fineberg. He went meta, and instructed ChatGPT to become its own prompt generator that is optimized for…you guessed it. ChatGPT.
“You are a prompt generation robot. You need to gather information about the users goals, objectives, examples of the preferred output, and other relevant context. The prompt should include all of the necessary information that was provided to you. Ask follow up questions to the user until you have confidence you can produce a perfect prompt. Your return should be formatted clearly, be optimized for chatgpt interactions, and begin with “You are an expert {profession}...”. Start by asking the user the goals, desired output, and any additional information you may need.”
SAVE THIS. Right now. Open up a file, write it on a sticky, put it in your diary. I don’t care, just take a moment to read it and put it somewhere easy to access.
Copy/Paste that into a chat, and answer the questions ChatGPT asks you until it has generated the desired prompt. You may need to rework it a bit, before starting a new chat with it, but it’ll give you something to play around with. See below for an example.
You: I would like to write a story about a pokemon trainer from pallet town who only has a magikarp, but is able to become the greatest pokemon trainer in the world!
ChatGPT: Great! Here's a prompt to help you write a story about a Pokémon trainer from Pallet Town who only has a Magikarp, but becomes the greatest Pokémon trainer in the world:
Act as an expert fiction writer and create a compelling story about a Pokémon trainer from Pallet Town who only has a Magikarp but is able to become the greatest Pokémon trainer in the world.
Introduce the protagonist: Begin by introducing the protagonist and their backstory. Explain how they ended up with only a Magikarp, and what their motivations are for becoming a great Pokémon trainer.
Create challenges and obstacles: Develop challenges and obstacles that the protagonist will face on their journey to become the greatest Pokémon trainer. For example, they could face an opponent who has a powerful team of Pokémon, or encounter situations that require them to think creatively in order to succeed.
Show the protagonist's growth: Throughout the story, show how the protagonist grows and changes. Perhaps they learn new techniques for training their Magikarp or develop a deeper understanding of their Pokémon's abilities. The protagonist's growth should be tied to the challenges they face, and should be a reflection of their hard work and dedication.
Introduce other characters: Along the way, introduce other characters who play a role in the protagonist's journey. This could include other trainers, gym leaders, or even rivals. Each character should have their own motivations and backstory that tie into the larger narrative.
Bring the story to a satisfying conclusion: As the protagonist overcomes challenges and grows, they should ultimately become the greatest Pokémon trainer in the world. Bring the story to a satisfying conclusion that highlights their success and showcases the skills and abilities they've developed along the way.
Remember to include exciting action scenes, emotionally resonant moments, and vivid descriptions of the Pokémon and their battles to keep your readers engaged throughout the story. Good luck, and have fun creating your story!
Now just plug the output into a new chat and voila! I’ve included the full story after the conclusion in case you’re interested in reading it.
Keep at it, young padawan
It's important to remember that prompt engineering is an ongoing process that requires testing and refinement over time, even with the prompt generator shortcut. The more you use ChatGPT and experiment with different prompts, the better you will become at crafting effective prompts and generating high-quality responses. Don't be afraid to try new things and iterate on your prompts as you go…or even asking ChatGPT itself to help you improve in the skill.
In conclusion, prompt engineering is a key aspect of using ChatGPT effectively, and can help you generate more accurate and helpful responses. By understanding the basics of prompt engineering, exploring different types of prompts, and fine-tuning your prompts over time, you can improve the quality of your ChatGPT interactions and get the most out of this powerful tool.
If you’re looking for more, I would highly recommend the article A Generic Framework for ChatGPT Prompt Engineering.
We hope you enjoyed this blog post. Be sure to follow us on LinkedIn, tiktok (@synapticlabs), our instagram (@synapticlabs). Don’t forget to check out our podcast "A Chat with ChatGPT", and check back frequently for more ChatGPT learning.
This blog was written in partnership with ChatGPT.
The Story of Maxicarp
In the small town of Pallet, a young boy named Max had always dreamed of becoming a Pokemon trainer. He spent hours watching televised Pokemon battles, studying moves, and creating strategies in his head. But he faced a significant obstacle - he didn't have a strong Pokemon to start with.
Max's father, a retired Pokemon trainer, had given him a Magikarp, a weak and useless Pokemon, to start his journey. Max was initially disappointed, but his determination didn't waver. He decided to make Magikarp the best Pokemon out there and set out on his journey.
Max and Magikarp traveled from town to town, participating in Pokemon battles. Despite his weak Pokemon, Max's strategies and determination helped him win battles against many trainers. He only had one move, Splash, but Max knew how to use it well. He would create distractions, lure his opponents into traps, and then use Splash to catch them off guard.
Max and Magikarp's unlikely victories soon became the talk of the town. People started talking about the boy who defeated others with a useless Pokemon. News of Max's legendary Magikarp began to spread to other towns, and other trainers started taking notice.
One day, Max received an invitation to participate in a Pokemon tournament organized by the Pokemon League. It was a huge opportunity for him and Magikarp to show their skills to the world.
Max and Magikarp made their way to the tournament, where they faced off against some of the strongest Pokemon trainers. The battles were intense, but Max's clever strategies and Magikarp's unexpected moves allowed them to make it to the final round.
In the final round, Max faced off against the strongest trainer in the tournament, whose team was composed of powerful and intimidating Pokemon. The odds were against Max, but he refused to give up. He believed in himself and in Magikarp.
The final battle began, and both trainers called out their Pokemon. The opponent's Pokemon quickly moved to attack, but Max was ready. He used his clever tactics to avoid their moves and then, in a sudden move, Magikarp used its Splash attack to knock out the opposing Pokemon.
The crowd erupted in cheers as Max and Magikarp were declared the winners of the tournament. Max had achieved the impossible, and his Magikarp had become a legend in the Pokemon world.
Max and Magikarp returned to Pallet Town as heroes. The town celebrated their victory, and Max's father was proud of his son's achievement. Max realized that anything is possible with determination and hard work, and he had proven that even the weakest Pokemon could achieve greatness.
From that day forward, Max and Magikarp's story became a source of inspiration for trainers everywhere. And so, the tale of the boy and his Magikarp lived on, a story of an unlikely victory against all odds.
Don’t forget to check out our podcast "A Chat with ChatGPT" for more information, or follow us on instagram (@synapticlabs) and tiktok (@chattingwithchatgpt).
This article was written in partnership with ChatGPT.
