To get the most out of this book

This book aims to provide a solid theoretical foundation of what LLMs are, their architecture, and why they are revolutionizing the field of AI. It adopts a hands-on approach, providing you with a step-by-step guide to implementing LLMs-powered apps for specific tasks and using powerful frameworks like LangChain. Furthermore, each example will showcase the usage of a different LLM, so that you can appreciate their differentiators and when to use the proper model for a given task.

Overall, the book combines theoretical concepts with practical applications, making it an ideal resource for anyone who wants to gain a solid foundation in LLMs and their applications in NLP. The following pre-requisites will help you to get the most out of this book:

• A basic understanding of the math behind neural networks (linear algebra, neurons and parameters, and loss functions)
• A basic understanding of ML concepts, such as training and test sets, evaluation metrics, and NLP
• A basic understanding of Python

Download the example code files

The code bundle for the book is hosted on GitHub at https://github.com/PacktPublishing/Building-LLM-Powered-Applications. We also have other code bundles from our rich catalog of books and videos available at https://github.com/PacktPublishing/. Check them out!

Download the color images

We also provide a PDF file that has color images of the screenshots/diagrams used in this book. You can download it here: https://packt.link/gbp/9781835462317.

Conventions used

There are a number of text conventions used throughout this book.

CodeInText: Indicates code words in text, database table names, folder names, filenames, file extensions, pathnames, dummy URLs, user input, and Twitter handles. For example: “I set the two variables system_message and instructions.”

A block of code is set as follows:

[default]
$pip install openai == 0.28
import os
import openai
openai.api_key = os.environment.get('OPENAI_API_KEY')
response = openai.ChatCompletion.create(
    model="gpt-35-turbo", # engine = "deployment_name".
    messages=[
        {"role": "system", "content": system_message},
        {"role": "user", "content": instructions},
    ]
)

Any command-line input or output is written as follows:

{'text': "Terrible movie. Nuff Said.[…]
 'label': 0}

Bold: Indicates a new term, an important word, or words that you see on the screen. For instance, words in menus or dialog boxes appear in the text like this. For example: “[…] he found that repeating the main instruction at the end of the prompt can help the model to overcome its inner recency bias.”