INDEX

Explanations

HTML

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ero

-0.80

 HTML

-0.69

AndEndTag

-0.67

erv

-0.59

HTML

-0.57

Personendaten

-0.57

 unknownFields

-0.56

ên

-0.56

su

-0.55

poster

-0.54

POSITIVE LOGITS

 itſelf

0.67

 GenerationType

0.65

 Adventures

0.62

 themſelves

0.59

ChildScrollView

0.58

 ſeveral

0.58

enumi

0.57

 Negra

0.57

 Diſ

0.57

Saiba

0.54

Activations Density 0.243%