INDEX

Explanations

list items or options

This neuron detects legal‐style numeric and parenthetical section/subsection markers (e.g. “16,” “(a),” “(b),” “Sec. 33.015”).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

️⃣

0.57

чной

0.56

 voorkomen

0.55

সম

0.54

Lucene

0.54

Bib

0.54

BorderSize

0.53

鞄

0.53

memset

0.53

閲覧

0.53

POSITIVE LOGITS

III

0.71

II

0.69

IV

0.64

 VIII

0.61

Ⅲ

0.57

iiv

0.57

VII

0.56

ii

0.54

II

0.54

XIV

0.53

Activations Density 0.047%