Back to Search Start Over

Scaling Laws for Neural Language Models

Authors :
Kaplan, Jared
McCandlish, Sam
Henighan, Tom
Brown, Tom B.
Chess, Benjamin
Child, Rewon
Gray, Scott
Radford, Alec
Wu, Jeffrey
Amodei, Dario
Publication Year :
2020
Publisher :
arXiv, 2020.

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.<br />Comment: 19 pages, 15 figures

Details

Database :
OpenAIRE
Accession number :
edsair.doi.dedup.....9a0fce293ba46e2c5dc5ac8b19a68892
Full Text :
https://doi.org/10.48550/arxiv.2001.08361