Machine learning
How large language models work
Tokens, embeddings, attention and the transformer, pretraining and fine-tuning. Enough to reason about what a model can and can't do.
About 15 hours of study.
Where to learn it
Read
The Illustrated Transformer
Jay Alammar
The whole post
The clearest picture of attention there is.
Section not checked yet
Watch
Neural Networks: Zero to Hero
Andrej Karpathy
Let's build GPT, then Intro to large language models
Section not checked yet
Practise
No practice set picked yet.
Suggest oneOther options
- Hugging Face LLM Course (read)
Chapters 1–2
Which roles ask for it
- AI / LLM engineer to get hired
Before and after
How large language models work