INDEX

Explanations

a single perfect

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

 Appearing

1.44

 Üniversitesi

1.43

 Evam

1.41

 Вул

1.34

 Wochen

1.32

 Stationery

1.31

 Hamburger

1.30

 Seamless

1.30

 Cavaliers

1.29

Cra

1.28

POSITIVE LOGITS

 conflict

0.94

its

0.85

 exist

0.80

 heter

0.79

 contain

0.79

 region

0.78

 offen

0.77

ുണ്ട്

0.76

 which

0.75

 ഉണ്ട്

0.75

Activations Density 0.090%