Basic Terminologies

AI, or Artificial Intelligence, refers to the field of computer science that focuses on creating intelligent systems capable of mimicking human thinking, learning, and understanding. The goal of AI is to develop machines that can perform tasks like writing, creating content, solving complex problems, drawing, coding, and programming.

NLP, or Natural Language Processing, is a subfield of AI that involves training computers to understand and process human language. Through NLP, computers can comprehend questions and provide relevant responses. It is NLP that enables applications like Siri, which can answer queries about the weather, and email filters that identify and block spam messages. NLP can be likened to trying to understand a language from another planet—it involves deciphering and comprehending human language patterns.

GPT, or Generative Pre-trained Transformer, is an NLP AI model widely used in various AI applications. In AI, we train computers to perform specific tasks, and the resulting output is referred to as an AI model. GPT is an NLP model specifically trained to understand human language. Different versions of GPT, such as GPT-2, GPT-3, GPT-3.5, and GPT-4, exist, with each version offering improved capabilities. For instance, ChatGPT employs GPT-3 and GPT-3.5 models.

LLM stands for Large Language Model and is a term frequently used in prompt engineering. It refers to language models, like GPT-3 or GPT-3.5, that have an extensive number of parameters. GPT serves as an example of an LLM. Large Language Models are trained on vast amounts of text data, enabling them to generate highly human-like text output.

Parameters in AI models are adjustable settings or "knobs" that determine the behavior and performance of the model. When we say that GPT-3 has 175 billion parameters, it means there are 175 billion individual settings that can be adjusted to enhance the model's performance in various language-related tasks. Parameters can be likened to puzzle pieces—having a greater number of pieces increases the chances of solving a puzzle correctly. Similarly, GPT-3's 175 billion parameters provide it with numerous "pieces" to solve language puzzles effectively, contributing to its ability to generate coherent and contextually appropriate text.

Understanding these basic terminologies is essential for prompt engineering, the process of designing and refining prompts or instructions to achieve the desired output from AI models. Parameters, NLP, GPT, and LLMs form the foundation of prompt engineering and enable the development of advanced AI systems capable of understanding and generating human-like text.