Learn to use AI

# Scaling

Scaling is a unit of 7 lessons in Rabbithole's Learn to use AI course, at levels 1 to 2. The same lessons are also in Become an AI engineer as “Scaling, plainly”.

## Lessons

1. Bigger models: what size buys you · Level 1
2. Pretraining: where an AI's knowledge comes from · Level 2
3. Sample efficiency: how little data it takes to learn · Level 2
4. Training compute: what it costs to build an AI · Level 2
5. Double descent: when getting worse comes before getting better · Level 2
6. Scaling laws: how AI gets better as it grows · Level 2
7. Emergent abilities: skills that appear as models grow · Level 2

## Sources

These lessons draw on 24 sources. The most used:

- [Scaling Laws for Neural Language Models](https://arxiv.org/abs/2001.08361)
- [Emergent Abilities of Large Language Models](https://arxiv.org/abs/2206.07682)
- [International AI Safety Report 2026](https://internationalaisafetyreport.org/sites/default/files/2026-02/international-ai-safety-report-2026.pdf)
- [Stanford CS25 V4 — Jason Wei and Hyung Won Chung of OpenAI](https://www.youtube.com/watch?v=3gb-ZkVRemQ)
- [How I use LLMs](https://www.youtube.com/watch?v=EWvNQjAaOHw)
- [Are Emergent Abilities of Large Language Models a Mirage?](https://arxiv.org/abs/2304.15004)
- [Deep Double Descent: Where Bigger Models and More Data Hurt](https://arxiv.org/abs/1912.02292)
- [Ilya Sutskever — moving from the age of scaling to the age of research](https://www.youtube.com/watch?v=aR20FWCCjAs)

[All units in Learn to use AI](https://rabbithole.pro/courses/learn-ai)

Source page: https://rabbithole.pro/courses/learn-ai/scaling