INDEX

Explanations

mathematical variable x

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

positive

-0.79

lope

-0.78

aute

-0.77

haupt

-0.75

neaux

-0.74

界

-0.73

想起

-0.72

ROSS

-0.72

прави

-0.71

 attackers

-0.71

POSITIVE LOGITS

 location

1.17

1.07

Locatie

0.98

Lieu

0.94

 locations

0.94

 Diarrhea

0.93

 위치

0.85

 center

0.84

 posición

0.82

Activations Density 0.026%