🛠️ Modifying the Text Generation Function – Live Coding with Sebastian Raschka (Chapter 5.3.3)

In this practical live-coding session, ML expert @SebastianRaschka dives into Chapter 5.3.3: Modifying the Text Generation Function from his book Build a Large Language Model (From Scratch). Now that decoding strategies like temperature scaling and top-k sampling have been introduced, it’s time to integrate them directly into the model’s text generation pipeline.

0:00 - Combining Temperature and Top-K Sampling
1:00 - Updating and Refactoring the Generation Function
3:01 - End-of-Sequence Token and Early Stopping
4:10 - Implementation Details and Testing
8:28 - Discussion: Why Output Quality is Still Random
9:46 - Practical Tips and Model Card Suggestions

📘 About the Book
Build a Large Language Model (From Scratch) is a practical and eminently-satisfying hands-on journey into the foundations of generative AI. Without relying on any existing LLM libraries, you’ll code a base model, evolve it into a text classifier, and ultimately create a chatbot that can follow your conversational instructions. And you’ll really understand it because you built it yourself!

💡 Perfect for ML engineers, AI developers, and NLP practitioners who want to experiment with custom generation strategies and build truly controllable LLMs.

🔗 Get the Book: https://hubs.la/Q03l0mSf0

📺 Subscribe to Manning Publications for expert-led tutorials, live coding series, and full transformer builds with Sebastian Raschka.

#SebastianRaschka #TextGeneration #LLM #GPT #DecodingStrategies #Transformers #PyTorch #DeepLearning #MachineLearning #NLP #ManningPublications #LiveCoding

Видео 🛠️ Modifying the Text Generation Function – Live Coding with Sebastian Raschka (Chapter 5.3.3) канала Manning Publications