Learn AI · 6 lessons · Free, live now
Under the Hood
Six lessons that open the box without the math. How words become a map of meaning, how a model decides which words matter, why the same prompt gives different answers, when fine-tuning is the right fix, what open weights means, and how models get scored.
When you finish, you can
Explain embeddings, attention, and temperature in plain English, tell prompting, retrieval, and fine-tuning apart, weigh open against closed models, and read a benchmark claim with a skeptical eye.
The lessons
- 01Embeddings: meaning as a mapHow words and sentences become lists of numbers, and why similar meanings land close together.
- 02Attention: what the model looks atHow a model decides which earlier words matter for the next one, shown on a single sentence.
- 03Temperature and samplingWhy the same prompt gives different answers, and what the creativity dial really changes.
- 04Prompting, retrieval, or fine-tuningThree ways to change what a model does, and how to tell which one a problem actually needs.
- 05Open and closed modelsWhat open weights means, why some models run on your own computer, and the privacy tradeoffs.
- 06Evals and benchmarksHow models get scored, what a leaderboard can and cannot tell you, and how to test one yourself.