INDEX

Explanations

UI control properties

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

 küpe

-0.87

familien

-0.87

 acondicionado

-0.85

 Dann

-0.85

familie

-0.84

はない

-0.84

hitung

-0.82

britannien

-0.82

 emulsi

-0.81

 İyi

-0.81

POSITIVE LOGITS

+=

1.06

0.99

ٍ

0.87

قی

0.87

 akin

0.86

XMLSchema

0.84

ु

0.81

mode

0.79

ianum

0.77

Mode

0.76

Activations Density 0.006%