Loading…
0%

Learn LLMs: From Zero to Transformer

What this course covers

From Zero to Transformer is a free, self-paced course that explains how large language models work, starting from no prior machine learning knowledge and ending at the mathematics of the original Transformer paper. It runs about six hours across ten modules, each with interactive labs and a checkpoint quiz. No installation and no account are required.

Who it is for

The course is written for technology leaders, analysts, engineers, and policy staff who need an accurate mental model of large language models rather than a surface summary. The early modules assume no background. The later modules assume comfort with vectors and matrix multiplication, though every step is explained.

Course syllabus

  1. Orientation (Start here, 10 minutes). How the labs, callouts, and checkpoints work.
  2. The Big Idea: Prediction Machines (Beginner, 35 minutes). Next-token prediction, what makes a model large, the stochastic parrot argument, and why raw prediction alone is not enough.
  3. A Brief History and the Landscape (Beginner, 30 minutes). From ELIZA to Transformers, the main model families, and how AI, machine learning, and large language models relate.
  4. Words to Numbers: Tokens and Embeddings (Beginner, 40 minutes). Tokenization, embeddings, vector arithmetic, and similarity measured by the dot product.
  5. How LLMs Learn: Training, Fine-tuning and RLHF (Intermediate, 45 minutes). Pre-training, the scale of compute involved, fine-tuning a base model into an assistant, and reinforcement learning from human feedback.
  6. Inside the Transformer I: Architecture and Data Flow (Advanced, 45 minutes). What the letters in GPT stand for, and how data moves through the stack.
  7. Inside the Transformer II: Attention (Advanced, 50 minutes). Self-attention and cross-attention, and the masking that stops a model reading ahead.
  8. The Original Paper and the Math (Expert, 40 minutes). A guided reading of Attention Is All You Need with the equations worked through.
  9. Capabilities, Limits and the Frontier (Expert, 35 minutes). Context windows, mixture-of-experts, multimodality, and knowledge distillation.
  10. Capstone: Exam, Glossary and Resources (Finish, 20 minutes). A final exam, a glossary of terms, and where to go next.

For shorter explanations of individual ideas, see the twelve AI concepts every tech leader should know. For the applied side, see Learn Agentic AI and the AI project portfolio.

About This Course

From Zero to Transformer is a free, interactive course on how large language models work, written by Shahzad Asghar. It takes you from the core idea of next-token prediction through tokenization, embeddings, training and RLHF, and into the Transformer architecture itself: the data flow, the attention mechanism, and the mathematics of the original "Attention Is All You Need" paper. Each module ends with a checkpoint, and hands-on labs let you experiment with the concepts directly in the browser. No mathematics or programming background is required to start.

Course Syllabus

  1. Orientation — how the course and interactive labs work
  2. The Big Idea: Prediction Machines — next-token prediction, the stochastic parrot, and what makes a language model large
  3. A Brief History and the Landscape — from ELIZA to Transformers, the model families, and AI vs ML vs LLM
  4. Words to Numbers: Tokens and Embeddings — tokenization, embeddings, word arithmetic, and vector databases
  5. How LLMs Learn — pre-training, the scale of compute, fine-tuning into an assistant, and RLHF
  6. Inside the Transformer I: Architecture and Data Flow — what G-P-T means, weights versus data, logits, softmax, and temperature
  7. Inside the Transformer II: Attention — queries, keys and values, the attention pattern, and multi-head attention
  8. The Original Paper and the Math — positional encoding, the attention formula, and the full Transformer block
  9. Capabilities, Limits and the Frontier — hallucination and how to fight it, and where the field is heading

This course is part of the free AI Learning Hub. For a visual companion, see LLM Architecture Explained; to go from models to autonomous systems, continue with Learn Agentic AI.