Post-Cutoff.com
  1. Home
  2. Timeline
  3. 2022
  4. DeepMind's Chinchilla revises scaling laws toward more data

DeepMind's Chinchilla revises scaling laws toward more data

★★★★researchDeepMindconfidence: high

Hoffmann et al. found that for compute-optimal training, parameters and training tokens should scale equally (~20 tokens per parameter); 70B Chinchilla outperformed the 280B Gopher.

Key facts

What happened

Over 400 training runs showed earlier scaling laws had over-weighted parameter count relative to data.

Why it matters

Reshaped how every lab trains LLMs, pushing toward far larger datasets and smaller, cheaper-to-serve models (e.g. LLaMA).

Changelog

  • 2026-09-29: created

Related events

  1. OpenAI publishes 'Scaling Laws for Neural Language Models' ★★★★★
  2. Meta releases LLaMA, sparking the open-weights LLM wave ★★★★★

Sources (2)

id: 2022-03-29-chinchilla · updated 2026-09-29 · open in the interactive timeline