INDEX

Explanations

RH Bill, Filipino children, conditions under

The neuron fires most strongly on tokens that appear in official or legal references—proper names, titles, sections/articles, dates, and other formal entities.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 featuring

-0.82

 waż

-0.77

摔

-0.76

intersect

-0.75

Featuring

-0.75

ッ

-0.73

 converting

-0.73

rickson

-0.72

PropertyName

-0.71

ENES

-0.70

POSITIVE LOGITS

 peasant

0.80

rights

0.72

 nạ

0.71

landa

0.71

 peasants

0.70

 broadest

0.70

 czerwony

0.70

Bip

0.69

ngo

0.69

 clutches

0.68

Activations Density 0.007%