AI 脉搏今日 +56
21/33 源在线

Scaling Laws, Carefully

摘要 · Scaling laws are one of the most critical empirical findings in deep learning. The observation is simple in form: the training loss $L$ decreases predictably as we scale up model size $N$, dataset size $D$, and compute $C$, following a power-law curve, which appears as a straight line on a log-log p
阅读原文 · Lilian Weng
政策监管

Scaling laws are one of the most critical empirical findings in deep learning. The observation is simple in form: the training loss $L$ decreases predictably as we scale up model size $N$, dataset size $D$, and compute $C$, following a power-law curve, which appears as a straight line on a log-log p

原文:https://lilianweng.github.io/posts/2026-06-24-scaling-laws/

0 条评论 · 观点来自社区

评论区 0 条讨论

的身份发言
还没有评论,来抢沙发。