AI Foundationspredict · compress · act

concepts → Property

Property

Compute-optimal

timeless3 connectionsdraft

the best split of a fixed compute budget between model size and data

Where it sits

The prism has six jobs across and eight layers down. Its primary cell ismeasure × L4. Hatched cells cannot exist — a GPU does not learn, an institution does not infer.

What must come first — and what it unlocks

Left to right is reading order, derived from the prerequisite_of edges. Nothing here is hand-ordered: the diagram is the graph.

Scaling lawCompute-optimalChinchilla

Before it: Scaling law

Unlocks: Chinchilla

What kind of thing it is — and what it is made of

It is part of: Scaling law.

Where to read it

The chapter that introduces it, and any chapter that uses it again.

27More Is Differentact 6 · The Attention Revolution

Where it comes from

paperTraining Compute-Optimal Large Language ModelsJordan Hoffmann, et al. · 2022

Every connection

All 3 edges touching this node, grouped by relation family — the sections above are highlights from this list. Colours match the relation families inthe atlas.

Structure · 1
is part ofScaling lawResult
Order · 2
requiresScaling lawResult
must come beforeChinchillaResult

This page is a projection of one node in src/data/concepts.ts. It has no prose file of its own — 230 declared edges produce all 236 of these pages. Edit an edge and both endpoints change.