Oreilly – Code with the Author of Build an LLM (From Scratch) 2025-5

Code with the Author of Build an LLM (From Scratch), Master the inner workings of how large language models like GPT really work with hands-on coding sessions led by bestselling author Sebastian Raschka. These companion videos to Build a Large Language Model from Scratch walk you through real-world implementation, with each session ending in a “test yourself” challenge to solidify your skills and deepen your understanding.

What you’ll learn

• Python Environment Setup
• Foundations to Build a Large Language Model (From Scratch)
• Tokenizing text
• Converting tokens into token IDs
• Adding special context tokens
• Byte pair encoding
• Data sampling with a sliding window
• Creating token embeddings
• Encoding word positions
• Computing the attention weights step by step
• Implementing a compact self-attention Python class
• Applying a causal attention mask
• Masking additional attention weights with dropout
• Implementing a compact causal self-attention class
• Stacking multiple single-head attention layers
• Implementing multi-head attention with weight splits

Specificatoin of Code with the Author of Build an LLM (From Scratch)

• Publisher : Oreilly
• Teacher : Sebastian Raschka
• Language : English
• Level : All Levels
• Number of Course : 54
• Duration : 18 hours and 18 minutes

Content of Code with the Author of Build an LLM (From Scratch)

Pictures

Sample Clip

Installation Guide

Extract the files and watch with your favorite player

Subtitle : English

Quality: 720p

Download Links

Download Part 1 – 1 GB

Download Part 2 – 1 GB

Download Part 3 – 1 GB

Download Part 4 – 486 MB

File size

3.47 GB