Unverified50% confidenceFactExact time
参数量越大模型能同时追踪的变量名、函数签名和调用关系越多，代码生成错误率与模型参数量之间存在幂律关系
1
Sources
50%
Confidence
Long-term
Relevance
7/13/2026
First Seen
Sources
Ornith 1.0 九B模型本地实测：塔防游戏暴露小模型编程瓶颈
Related Claims
Unverified提示词与模型版本存在深度绑定关系，不同版本模型在RLHF训练数据、指令微调策略上的差异会导致相同提示词产生不同输出，这种现象被称为「提示词漂移」71% similarUnverifiedBecause large language models are based on probabilistic sampling, even if instructions are written in the context, the model may still forget or skip certain instructions in complex reasoning chains, which is known as instruction-following instability.70% similarUnverifiedThe number of model parameters (e.g., 120B meaning 120 billion parameters) is a key indicator of large language model scale; more parameters generally means greater knowledge capacity and reasoning ability, but higher inference computation costs and slower response speeds.69% similarUnverified在纯Agent场景下，模型的指令遵循能力往往比参数量更重要69% similarUnverifiedScaling laws reveal a near power-law relationship between model loss and the number of parameters, data volume, and compute69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/502661API
curl https://kongchang.com/api/v1/knowledge/claims/502661MCP
get_claim(id=502661)