Scaling Laws for Neural Language Models

TLDR

The study investigates empirical scaling laws of language model cross‑entropy loss. The authors derive simple equations linking overfitting and training speed to model and dataset size. The loss follows a power‑law with model size, dataset size, and compute, while architectural changes have little effect; these relations enable optimal compute allocation, showing that large models are more sample‑efficient and that compute‑efficient training uses very large models on modest data with early stopping.

Abstract

We study empirical scaling laws for language model performance on the cross-entropy loss. The loss scales as a power-law with model size, dataset size, and the amount of compute used for training, with some trends spanning more than seven orders of magnitude. Other architectural details such as network width or depth have minimal effects within a wide range. Simple equations govern the dependence of overfitting on model/dataset size and the dependence of training speed on model size. These relationships allow us to determine the optimal allocation of a fixed compute budget. Larger models are significantly more sample-efficient, such that optimally compute-efficient training involves training very large models on a relatively modest amount of data and stopping significantly before convergence.