Unverified60% confidenceFactExact time
The Scaling Law paper by Kaplan et al. in 2020 revealed that, under a fixed compute budget, there is an optimal ratio between model parameter count and the number of training tokens (the Chinchilla optimal).
2
Sources
60%
Confidence
Long-term
Relevance
7/16/2026
First Seen
Sources
Related Claims
VerifiedIn 2020, Kaplan et al. at OpenAI revealed that model performance follows predictable power-law relationships with parameter count, training data volume, and compute budget.80% similarUnverifiedThe core finding of the Chinchilla scaling law is that under a fixed compute budget, simultaneously increasing model parameters and training data volume is required to achieve optimal performance.77% similarVerifiedThe scaling laws established by Kaplan et al. in 2020 indicate that model performance follows a power-law relationship with parameter count, data volume, and compute.74% similarUnverified2020年Kaplan等人提出原始Scaling Law，倾向于在固定算力下优先扩大模型参数量74% similarUnverified2020年Kaplan等人提出的原始Scaling Law倾向于在固定算力下优先扩大模型参数量，导致GPT-3等模型相对训练数据量参数偏大72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/530834API
curl https://kongchang.com/api/v1/knowledge/claims/530834MCP
get_claim(id=530834)