Unverified50% confidenceFactExact time
Negative log-likelihood (NLL) refers to the negative average log of the probability a model assigns to each token in a given text; a lower value indicates the model finds the text more plausible.
1
Sources
50%
Confidence
Long-term
Relevance
11/6/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe sum of the negative log probabilities of each token output by a language model is an estimate of the encoding length of the sequence, which is the origin of the perplexity metric.77% similarUnverifiedNegative log-odds is defined as the negative of the difference between the log probability of the correct class and the log probability of the incorrect class.73% similarUnverifiedThe model's output tokens are zero, because it only returns probabilities rather than generating text.70% similarUnverifiedTwo's complement represents negative numbers by flipping every bit of the corresponding positive number and adding 1.68% similarUnverified困惑度是测试集上每个token平均负对数似然的指数，值越低代表模型对语料的预测越准确68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1142392API
curl https://kongchang.com/api/v1/knowledge/claims/1142392MCP
get_claim(id=1142392)