Unverified50% confidenceFactExact time
温度参数控制采样的随机程度，核采样（top-p）从概率最高的token子集中随机选取
1
Sources
50%
Confidence
Long-term
Relevance
7/16/2026
First Seen
Sources
Related Claims
Verified温度（temperature）参数控制采样过程，温度越高输出越随机多样，温度越低越趋向最高概率的选择84% similarUnverifiedtemperature 参数控制输出随机性，top_p 通过动态截断概率分布平衡输出多样性与质量79% similarUnverifiedA model's vocabulary typically contains 30,000 to 100,000 tokens, and the degree of randomness in sampling is controlled via the Temperature parameter.73% similarUnverifiedWhen testing citation behavior, temperature should be set to 0 or a low value with the top-p parameter fixed, and the same prompt should be run at least 20–50 times to use statistical distributions instead of single results.72% similarUnverified较高的温度参数会增加输出的随机性，在测试场景中往往需要将其调低以保证输出稳定性72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/530241API
curl https://kongchang.com/api/v1/knowledge/claims/530241MCP
get_claim(id=530241)