Unverified50% confidenceFactExact time
LLM在每个生成步骤输出的是词汇表上的完整概率分布，通常通过softmax函数计算
1
Sources
50%
Confidence
Long-term
Relevance
7/14/2026
First Seen
Sources
Prompt中一个词如何左右AI输出：可能性之井原理解析
Related Claims
UnverifiedThe core mechanism of large model inference is the continuous prediction of the probability of the next word, converting raw scores into a probability distribution summing to 1 via the Softmax function.79% similarUnverified模型对所有Token计算原始得分Logit后，通过Softmax函数将得分归一化为概率分布，所有Token的概率之和等于177% similarVerifiedLLMs generate output token by token in an autoregressive manner, with each word's selection based on probability sampling from the current context.76% similarPartially VerifiedLLMs use an autoregressive generation mechanism, sampling the next token at each step based solely on a probability distribution.76% similarUnverifiedLLMs are essentially next-token prediction machines based on statistical patterns, and their output is the text sequence with the highest probability given the context.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/509944API
curl https://kongchang.com/api/v1/knowledge/claims/509944MCP
get_claim(id=509944)