INDEX

Explanations

obligated

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

 Office

-0.85

Office

-0.66

werk

-0.65

ette

-0.64

ath

-0.63

ed

-0.60

 advance

-0.58

Advance

-0.58

Of

-0.57

 Advance

-0.57

POSITIVE LOGITS

Efq

0.92

 Jefus

0.84

 itſelf

0.83

 Majefty

0.82

 iſt

0.81

RectangleBorder

0.81

 myſelf

0.79

 poffe

0.77

ſelf

0.76

 leaſt

0.76

Activations Density 0.083%