thinQmaster Virtual testing partner Adaptive practice exams that find your weak spots and drill them. Every answer explained. From $69
Enroll

thinQmaster

Find your exam

Five questions free, before you decide.

Step 1 · choose a vendor

In one of our camps? Your practice exams come with the tuition — they are on your study page already.

AI & Machine Learning · Intermediate

Perplexity

A language-model metric measuring how well a model predicts text; lower perplexity means better prediction.

Definition

Perplexity is the exponential of the average cross-entropy per token. It reflects how surprised the model is by held-out text and is useful for comparing language models on the same tokenizer and data, but it does not directly measure helpfulness or factuality.

Exam tip

Lower perplexity is better; it is not a measure of truthfulness.

Related terms