INDEX

Explanations

Appellant and Appeal from

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

 généralement

-0.76

同士

-0.75

éndez

-0.75

 Fieber

-0.75

 лицен

-0.75

ifiers

-0.74

 Drittan

-0.74

Prés

-0.73

remacy

-0.73

 стоит

-0.73

POSITIVE LOGITS

 Relatives

0.81

瓣

0.71

ulario

0.71

usleep

0.70

ҫ

0.70

得点

0.68

 guid

0.68

 worse

0.68

langs

0.68

 quests

0.67

Activations Density 0.024%