Large Language Model

Everyone can see what a horse is.

– Benedykt Chmielowski, Nowe Ateny (1745)

For practical purposes, in this handbook, any AI model that can handle user prompts to produce human-like text and follow instructions is considered a “large language model.” This includes GPT, Llama, and any other models that may be developed in the future.

These models are typically trained on a large corpus of text data and can generate human-like text in response to user prompts.

In this handbook, we will discuss how to use these models, fine-tune them, evaluate their performance, and build applications around them.

We will not focus on how to create new Large Language Models nor on their internal architecture besides the basics.