Skip to content
Road to Intelligence

Concept · Chapter 8: The Rise of Large Language Models

Scaling Laws (Preview)

Must knowUnderstand12 minDifficulty

A language model's loss falls smoothly and predictably, as a power law, as its parameters, training data and compute grow, which turned 'make it bigger' into a plannable engineering decision.

The problem

Training a frontier model costs months and vast sums; teams needed to know in advance whether scaling up would pay off, and how to split a budget between model size and data.

The solution

Train many smaller models, fit how loss depends on parameters N, data D and compute C, and extrapolate.

The consequence

Labs could predict big-model loss from small runs and justify ever-larger investments; later work (Chinchilla) corrected the recommended balance toward much more data.

What was found

Kaplan and colleagues (2020) found that language-model cross-entropy loss falls as a power law in model size, dataset size and training compute, with some trends spanning more than seven orders of magnitude, while details such as width versus depth matter little within a wide range Established. A power law looks like

L(N)≈(NcN)αL(N) \approx \left(\frac{N_c}{N}\right)^{\alpha}

with a small exponent α\alpha (below 0.1 for parameters in their fits). Take logarithms and it becomes a straight line: every tenfold increase in parameters cuts the loss by the same fraction.

Tiny example. With α=0.076\alpha = 0.076, multiplying parameters by 10 multiplies the loss by 10−0.076≈0.8410^{-0.076} \approx 0.84: a 16% reduction, again and again, for each tenfold step. Steady, but each step costs ten times more.

Why it mattered

Predictability turned scale into planning. OpenAI reported predicting aspects of GPT-4's performance from models trained with no more than 1/1,000th of its compute Established. The training compute of a dense model is roughly C≈6NDC \approx 6ND floating-point operations, so a budget can be split between parameters N and tokens D in advance.

The Chinchilla study (2022) revised the recommended split: for a fixed compute budget, parameters and training tokens should grow in equal proportion, and a 70B model trained on 1.4 trillion tokens outperformed the 280B Gopher Established. That is about 20 tokens per parameter, a ratio now often quoted as a rule of thumb. Chapter 9 works through the numbers.

What to remember

  • Loss ≈ power law in N, D and C, with trends over more than seven orders of magnitude (Kaplan et al. 2020).
  • On a log–log plot, a power law is a straight line: steady returns for each multiplication of resources.
  • Chinchilla (2022): scale parameters and tokens together; about 20 tokens per parameter in its own case.
  • Scaling laws predict loss, not specific abilities, and they describe trends, not guarantees.

Key papers

Essential

Scaling Laws for Neural Language Models

Jared Kaplan, Sam McCandlish et al. · 2020

Found that language-model loss falls as a smooth power law in parameters, data and compute — making model scale something you could plan.

~1 h readarXiv:2001.08361✓ verified 2026-09-26
Essential

Training Compute-Optimal Large Language Models

Jordan Hoffmann, Sebastian Borgeaud et al. · 2022 · NeurIPS 2022

Showed that many large models were undertrained: for a fixed compute budget, parameters and training tokens should grow roughly in proportion.

~1 h readarXiv:2203.15556✓ verified 2026-09-26
Important

GPT-4 Technical Report

OpenAI et al. · 2023

Documented a large jump in capability, including human-level scores on many professional and academic exams, and marked the point where frontier labs stopped disclosing model size, data and training details.

How to read it: Note what the report does not contain: architecture, parameter count, data and compute are all withheld.

~45 min readarXiv:2303.08774✓ verified 2026-09-26