---
title: "Scaling Laws for Neural Language Models"
description: "The scaling-laws paper turns model size, data, and compute into a quantitative tradeoff. It is useful when you want to reason about progress curves instead of treating bigger training runs as folklore."
canonical_url: "https://fanout.sh/daily/2026-07-17-scaling-laws"
md_url: "https://fanout.sh/daily/2026-07-17-scaling-laws.md"
last_updated: "2026-07-17"
access: "public"
---

# Scaling Laws for Neural Language Models

The scaling-laws paper turns model size, data, and compute into a quantitative tradeoff. It is useful when you want to reason about progress curves instead of treating bigger training runs as folklore.

## Paper details

- Authors: Kaplan et al.

- Venue: arXiv 2020

- Track: ML Math

- Difficulty: Medium

- Reading time: 50 min

- Original paper: https://arxiv.org/abs/2001.08361

## What you will learn

- How loss trends with parameter count, dataset size, and compute budget.

- Why log-log plots are the natural language of empirical scaling.

- Where simple laws help and where newer regimes can break them.

## Continue exploring

[Explore Scaling Laws](https://fanout.sh/knowledge-graph?node=ai-research%3Aconcept%3Ascaling-laws): Open Fanout's scaling-laws node and connect the paper to model size, data budget, and LLM training decisions.

---
This representation contains public Fanout content only. Protected Pro lessons, account data, billing, checkout, and pricing are not included.

Browse the public content map: https://fanout.sh/sitemap.md