Unverified50% confidenceFactExact time
Chinchilla定律大致按每个参数分配20倍token的比例来安排算力
1
Sources
50%
Confidence
Long-term
Relevance
8/13/2026
First Seen
Sources
Related Claims
UnverifiedThe Chinchilla optimal scaling strategy proposed by DeepMind states that for a given compute budget, there is an optimal ratio of parameter count to data volume, approximately 1:20.63% similarVerifiedThe MoE architecture typically activates only 10%–20% of total parameters during each inference62% similarUnverified对于 198B 总参数的稀疏 MoE 模型，实际激活参数量可能仅为 20-40B 量级62% similarUnverified参数量在1亿以下的轻量模型经领域数据微调后可达95%以上意图分类准确率，单次推理延迟控制在5-20ms以内61% similarUnverified在35B参数模型中GEMM内核往往占据总计算时间的60-80%59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/742553API
curl https://kongchang.com/api/v1/knowledge/claims/742553MCP
get_claim(id=742553)