INDEX

Explanations

feature

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

@"/

-0.67

 ویکی‌پدیا

-0.65

triangleq

-0.49

 occhiali

-0.49

msgTypes

-0.48

(!__

-0.48

 LUMP

-0.47

providedIn

-0.47

 Kär

-0.47

Eksteraj

-0.45

POSITIVE LOGITS

 houſe

0.83

 Jefus

0.79

 Houſe

0.75

 purpoſe

0.74

 pleaſure

0.74

 viſ

0.73

niosek

0.69

 ſmall

0.68

 ſub

0.67

 reaſon

0.66

Activations Density 0.014%