July 25, 2023

What is Prompt Engineering? How to write an effective GPT-3.5 or GPT-4 prompt?

Discover the power of prompt engineering in AI, and the advancements in GPT-3.5 and GPT-4 prompts.

Rishika Shidling

What is Prompt Engineering? How to write an effective GPT-3.5 or GPT-4 prompts?

Prompt engineering has become a trending job in the world of artificial intelligence (AI). AI is a technology that allows machines to think and act like humans. Similarly, prompt engineering is a way of AI algorithms using data analysis. It is a natural language processing (NLP) concept that involves discovering inputs that yield desirable results. Artificial Intelligence and Data Analytics are the power to analyze and learn about large amounts of data from multiple sources and detect patterns to make future predictions.

Prompt engineering has become one of the most demanding professions in 2023. While there are concerns about the accuracy of the information it generates in response to the prompts given, businesses are moving quickly to incorporate LLMs into all manners of communication.  Large Language Model (LLM) AI is a term that refers to AI models that can generate natural language texts from large amounts of data. Large language models use deep neural networks, such as transformers, and can learn from billions of words and create texts on a range of topics.

Neural networks are also parameterized models that are learned with continuous optimization methods. In fact, neural networks can be viewed as more powerful versions of these simple models, with this power being achieved by combining the basic models into a comprehensive neural architecture. Some LLMs are referred to as foundation models, a term coined by the Stanford Institute for Human-Centerer Artificial Intelligence in 2021.

As it travels through the transformer neural network process, the LLM engages in deep learning. Using a self-attention mechanism, the transformer architecture enables the LLM to comprehend and recognize the interconnections and connections between words and concepts. In order to establish the relationship, that system can give a particular item (referred to as a token) a score, also known as a weight.

Human VS computer

Humans and computers are inherently suited to different types of tasks. For example, computing the cube root of a large number is very easy for a computer, but it is extremely difficult for humans. On the other hand, a task such as recognizing the objects in an image is a simple matter for a human but has traditionally been very difficult for an automated learning algorithm.

Only in recent years has deep learning shown accuracy on some of these tasks that exceed that of a human.

Humans often learn simple concepts first and then move to the complex. The training of a child is often created using such a curriculum in order to accelerate learning. The human brain is examined and used to model various artificial neural networks in artificial intelligence in order to create a computerized expert system. The brain is a natural organ of the human body that can make decisions on its own, but a computer is a machine that was created and that needs enough information and instructions to make a decision.

The most significant difference between the brain and a computer is that the human brain has the ability to make decisions on its own and it can store an infinite amount of information. In contrast, a computer has to be programmed to perform the functions and has a limited capacity to store data and information.

What is GPT-3.5 or GPT-4 prompt?

First, let’s understand what is GPT-3.5 or GPT-4 prompts. GPT-3 and GPT-4 predict text based on an input called a prompt. The best results, though, come from writing a prompt that is crystal clear and has a lot of context.

On November 30th, GPT-3.5 was released to the public along with ChatGPT, a refined version of GPT-3.5 that essentially functions as a general-purpose chatbot.

It debuted in front of the public with a demonstration of its conversational abilities on a variety of topics, including programming, TV scripts, and scientific ideas.

GPT-3.5 was trained on a combination of text and code by the end of 2021, according to Open AI, when comparing GPT-3 and GPT-3.5. It acquired this knowledge by ingesting a substantial amount of content from the web, such as hundreds of thousands of Wikipedia pages, social media posts, etc.

Unlike GPT-3.5, which focuses primarily on text, GPT-4 can analyze and comment on images and graphics.GPT-4 is ten times more advanced than GPT-3.5, which it replaced. This will enable the model to recognize subtleties and get a deeper understanding of the context, resulting in replies that are more accurate and consistent.

GPT-3.5 or GPT-4 which one is better?

GPT-4 is 10 times more advanced than GPT-3.5. While GPT-3.5 is quite capable of producing prose that is human-like, GPT-4 is even better at comprehending and producing many dialects and reacting to the emotions represented in the text.

For instance, GPT-4 can detect and delicately react to a user expressing melancholy or annoyance, making the conversation feel more intimate and sincere. While GPT-3.5 might have trouble making connections, GPT-4 can synthesize data from several sources to provide answers to complex questions.

For instance, GPT-4 can give a more thorough and nuanced response, referencing various research and sources, in response to a question regarding the relationship between the loss of bee numbers and the effect on world agriculture.

How to write an effective GPT-3.5 or GPT-4 prompt?

1. Write a better prompt: The secret to writing a better prompt is to use precise language and convey as much context as you can.

Basic prompt: "Write a poem about winter floods and spring."

Better prompt: "Write a poem in the style of Emily Brontë about floods and showers of spring,”

2. Mention details: You may mention things like:

Your desired topic, format, style, intended audience, text length, a list of the issues you want to be addressed, and, if applicable, the standpoint from which you want the text to be written.

3. Specify the length of response you want:

It's helpful to include a word count when creating GPT prompts so that you don't receive a 500-word answer when you were seeking a phrase (or vice versa). Even better, list a few acceptable lengths.

Follow these 3 most important strategies to create a better prompt for GPT 3.5 and GPT 4.


Numerous businesses have used prompt engineering and are now reaping the rewards in the form of greater productivity, enhanced customer satisfaction, and cost savings.

What is HuggingChat? Everything you need to know about this open-source AI chatbot
A compact analysis of HuggingChat.
Rishika Shidling
MiniGPT-4: Open-Source Model for Complex Vision-Language Tasks Like GPT-4
A pithy outlook on MiniGPT-4: an open-source model for complex vision-language tasks like GPT-4
Rishika Shidling
AutoGPT vs BabyAGI
A concise comparison between AutoGPT and BabyAGI.
Rishika Shidling