Unverified50% confidenceFactExact time
The Chinchilla scaling law states that compute-optimal training requires approximately 20 tokens per parameter.
1
Sources
50%
Confidence
Long-term
Relevance
11/3/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe classic Chinchilla scaling law points out that for each FLOPs budget, one should scan different model sizes and choose the optimal one, with the rule of thumb being that the number of training tokens is about 20 times the number of parameters.79% similarVerifiedIn 2022, DeepMind published the Chinchilla paper, which corrected the optimal training ratio and indicated that, under a given compute budget, the number of model parameters and the number of training tokens should maintain an approximate ratio of 1:2075% similarUnverified训练FLOPs可通过Chinchilla缩放定律估算，计算量约等于6乘以参数量乘以训练token数73% similarUnverifiedThe experiment trains a small MLP on a modular addition task modulo 97, with the training set making up only 35%, using full-batch training for a total of 20k steps.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1130349API
curl https://kongchang.com/api/v1/knowledge/claims/1130349MCP
get_claim(id=1130349)