INDEX

Explanations

URL fragments and codes

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

of

-1.79

 détaillée

-1.32

معلومات

-1.31

 rudimentary

-1.26

 ответствен

-1.26

 always

-1.25

 distraught

-1.24

any

-1.23

 yıllar

-1.23

 eventuali

-1.23

POSITIVE LOGITS

xvii

1.67

atized

1.59

{'

1.52

 they

1.52

 XVII

1.52

 февра

1.50

//{

1.50

INFINITY

1.50

cular

1.49

ensitive

1.48

Activations Density 0.048%