Unverified50% confidenceFactExact time
The LLaMA series adopts an over-training strategy, training a 7B-parameter model on 1T tokens, far exceeding the Chinchilla-optimal ratio of 140B tokens
1
Sources
50%
Confidence
Long-term
Relevance
9/11/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedAccording to the Chinchilla scaling law, the optimal training data volume for a 30B model is approximately 600B–1T tokens.69% similarUnverifiedThe 70-billion-parameter Chinchilla model, trained on 1.4 trillion tokens, outperformed Gopher, which has 280 billion parameters but was trained on only 300 billion tokens.63% similarUnverifiedLoRA使可训练参数量通常仅为全参微调的0.1%-1%，可在单张RTX 3090/4090（显存24GB）上微调7B乃至13B级别的模型63% similarUnverified叠加cv=5的5折交叉验证后，144种参数组合需训练144×5=720次模型58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/893366API
curl https://kongchang.com/api/v1/knowledge/claims/893366MCP
get_claim(id=893366)