Concepts

Overview

This concept glossary is designed to be a quick reference for foundational ideas in AI, with a focus on techniques for working with large language models (LLMs) and prompt engineering. Whether you’re new to AI or looking to deepen your understanding of specific terms, this glossary will help clarify the basics and highlight connections within the field.

Each entry provides essential information in layman’s terms, along with links to related resources and concepts for further exploration.

Concepts

AI

AI (Artificial Intelligence) is the field of creating machines or software that can perform tasks typically requiring human intelligence, like understanding language, recognizing images, or making decisions.

Machine Learning

Machine Learning (ML) is a method of training AI systems to learn from data, improving their performance on specific tasks without explicit programming.

Deep Learning

Deep Learning is a branch of machine learning that uses neural networks with many layers to analyze complex patterns in large datasets.

Generative AI

Generative AI refers to systems that create new content, such as text, images, or music, by learning from existing data.

LLM (Large Language Model)

Large Language Models (LLMs) are powerful AI models trained on massive text datasets to understand and generate human language.

Embedding

An embedding is a way of representing words or concepts as numbers to capture their meaning and relationships in a model.

Token

A token is a small piece of text, such as a word, subword, or character, that language models use as the building blocks for generating language.

Dense vs. Sparse Vectors

Dense and sparse vectors are two types of data representations, with dense vectors having mostly non-zero values and sparse vectors having mostly zero values.

Prompt

A prompt is an instruction given to a model, guiding it to generate specific types of responses.

Sampling

Sampling is a method of selecting the next word in text generation based on probability, adding randomness to the output.

Fine-Tuning

Fine-tuning is a method of further training an AI model on specific tasks or domains to improve its performance in targeted applications.

Few-Shot Learning

Few-shot learning is a technique where a model is trained to perform a task with very few examples, making it adaptable with minimal data.

Meta-Prompting

Meta-prompting is a strategy that uses prompts to shape or refine how other prompts are formulated, guiding AI output by prompting at a meta level.

Decoding

Decoding is how AI picks words to build its response, influencing how natural or accurate the output sounds.