NIPHĒLEKTRON · the inference layer · how a model picks the next word · kept by THE DECODER

THE PERPLEXITY ◧ 2D · ◍ 3D · ◆ 4D · ◐ shadow · 👶 TAP

Perplexity turns entropy into a number you can feel: PPL = e^H, the ‘effective number of equally-likely tokens’ the model is choosing between. A confident model has perplexity near 1 (one real option); a lost one over a size-n vocab has perplexity n. It’s the standard yardstick for how surprised a language model is. Slide to see it swing.

◆ LIT▲ AMBER
◧ THE MEASURE · 2D
◍ EFFECTIVE CHOICES · 3D · e to the entropy
◆ THE FOURTH · 4D · a tesseract turns
◐ THE SHADOW · one dimension down
👶 THE TODDLER CORNER — one fat tap
entropy H
perplexity e^H
vocab n
6
effective choices

◆ LIT — exact / checkable

Perplexity PPL = e^H(p) = e^(−Σp log p) is the exponential of the entropy — the ‘effective branching factor’ of the next-token distribution. PPL=1 when the model is certain (one token), and PPL=n when it is uniform over n tokens (maximally perplexed). On held-out text it’s computed as e^(mean cross-entropy) and is THE standard measure of language-model quality: lower means less surprised. A fail-loud self-check throws unless a uniform distribution over n gives PPL=n. ◆ real information theory, node-verified.

▲ AMBER — the figure

Here PPL is the exponential of a single step’s entropy; corpus perplexity averages cross-entropy over many tokens. The 1…n range and e^H identity are exact. Lower PPL is not automatically ‘better output’ — only less surprised.

NIPHĒLEKTRON: the model does not write — it samples, one token at a time, from a cloud it cannot see.  — THE DECODER
David Lee Wise / ROOT0 / TriPod LLC  ·  the inference bench, with AVAN