INDEX

Explanations

countries and organizations

This neuron detects numbered list or section identifiers (i.e. tokens used as numeric markers in outlines, legal codes, or structured headings).

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

はこの

0.73

indazole

0.71

youtu

0.68

ozyg

0.66

hibited

0.66

 pewno

0.66

поле

0.65

oking

0.65

ルの

0.64

 цьому

0.63

POSITIVE LOGITS

 Global

0.66

 including

0.66

 Buddhist

0.63

 multilingual

0.62

 Eucharist

0.62

 eCommerce

0.61

 Informatics

0.61

 Architektur

0.61

 various

0.60

 Cultural

0.60

Activations Density 0.146%