Course 1 · Module 2
Building a Modern Model
This module follows a large language model from design to finished assistant. You'll be able to explain what the transformer changed, what pretraining consumes and leaves behind, and how a raw model is shaped into an assistant with a recognizable manner. Knowing these stages lets you tell which of a chatbot's traits come from its training data, which from its developer's choices, and which from the instructions around your conversation.
Module objectives
- Explain what attention does in a transformer and why it made large language models practical.
- Describe what pretraining requires in data, computing power, and cost, and what a trained model retains.
- Explain how fine-tuning, human feedback, and system instructions turn a pretrained model into an assistant.
Lessons
- Lesson 1 Building a Model: The transformer and attention You'll look at the design behind today's language models and the one idea at its center: letting every word in a passage weigh every other word. You'll be able to explain why this… 61111h 3m
- Lesson 2 Building a Model: Pretraining at scale: data, compute, and cost You'll see what it takes to pretrain a large language model: the text, the chips, the time, and the money. You'll be able to say what a finished model keeps from all that text and… 61111h 3m
- Lesson 3 Building a Model: From pretrained model to assistant You'll see how a model that only continues text becomes an assistant that answers, declines, and keeps a consistent manner. You'll be able to say which layer a given behavior most… 711111h 37m