INDEX

Explanations

not followed by verb/adjective

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

 ใน

-1.58

髕

-1.55

 frança

-1.51

 gênero

-1.50

 artesanato

-1.48

いいのか

-1.48

"\

-1.46

 verwende

-1.43

蛱

-1.41

笵

-1.41

POSITIVE LOGITS

stva

1.42

ogues

1.36

ỗng

1.30

 черво

1.30

)：

1.27

kkor

1.26

 they

1.23

but

1.22

 there

1.21

 people

1.20

Activations Density 0.014%