Scaling Laws for Neural Language Models (2020)
研究论文
规模化
Kaplan et al.: loss falls as a power law in compute — the investment thesis, quantified.
您正在查看一条研究论文记录：一篇里程碑式的论文，以及它为该领域带来了什么。
This OpenAI paper measured smooth power-law relationships between model performance and parameters, data and compute — the curves that turned 'bigger is better' from folklore into planning tool, and ultimately into capex justification for everything on this site.
它如何与整个技术栈相连
本记录的关联链接一览——点击分组可直接跳转。
推出于
基础来源
来源
记录审阅截至 2026-01-15 — 查看 方法论.