Sign inSign in
ConceptAI & ML

Neural scaling laws

Language modeling performance follows a predictable power-law scale with model size in terms of either compute or parameters.
Watch it explained

Know first

Named as a prerequisite by the lectures that define this. The number is how many say so. A grey one has no page yet.

Compute budget1Language modeling1Model parameters1Neural networks1Power law1

Opens up

Concepts that name this one as a prerequisite.

Lecturers

2

Each in their own terms, from the easiest level.

  • Irina Rish

    Université de Montréal, MILA · Research

    Functional forms that describe how neural network systems scale with respect to resource factors like dataset size, model size, and compute.
    • 21:09Emergent Behaviors in Foundational Models

Connections

6

The lectures that define it or assume it, and the concepts next to it.

Concepts4

Lectures2

2 lecturers. Definitions are written by AI from each lecture’s transcript. A time with a play mark starts the lecture at the moment the lecturer says it.