INDEX

Explanations

German cities and regions

The neuron strongly activates on named entities—proper nouns such as organization names, place names, dates, and other capitalized or distinct entity mentions.

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

മ്മ

0.91

cription

0.70

每

0.68

0.67

و

0.67

sink

0.65

 presse

0.64

말로

0.64

 huid

0.64

pointe

0.63

POSITIVE LOGITS

 patties

0.84

 parameters

0.83

 chestnuts

0.80

 diapers

0.79

 параметры

0.78

 Gather

0.78

 subset

0.77

 fervor

0.77

EMBER

0.77

 kernels

0.77

Activations Density 0.000%