The paper reports that loss scales as a power law with model size, dataset size and training compute, with some trends spanning more than seven orders of magnitude.
One recorded claim. The claim panel gives its qualification, the source passages with their place in each source, and how it was checked.
What it connects
Scaling Laws for Neural Language Models → states its contribution → The paper reports that loss scales as a power law with model size, dataset size and training compute, with some trends spanning more than seven orders of magnitude.
Where it appears
No collection page holds this claim. It is listed with its records in the explorer.