INDEX

Explanations

observing how people are

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 justice

1.32

𝒸

1.22

po

1.22

pom

1.21

 Kardashians

1.20

т

1.19

 seriously

1.19

nic

1.18

 schade

1.16

 clearTimeout

1.16

POSITIVE LOGITS

iapkan

1.55

ت

1.51

ing

1.42

geoning

1.41

anço

1.39

ed

1.38

alco

1.33

ের

1.32

sunday

1.32

Config

1.31

Activations Density 0.003%