INDEX

Explanations

identification, welcome

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Coordenadas

-0.84

 even

-0.78

いかがでしたか

-0.75

Hvor

-0.73

 Valentino

-0.72

衬衫

-0.71

how

-0.71

 orde

-0.71

 abaixo

-0.70

Billi

-0.70

POSITIVE LOGITS

 hundred

0.94

 Partager

0.82

 paragraph

0.81

 centaines

0.78

 zusätz

0.78

 varargin

0.78

 StatefulWidget

0.77

 نقطة

0.77

intra

0.77

̵

0.76

Activations Density 0.011%