INDEX

Explanations

programming code types

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

 одним

-1.56

 vulga

-1.33

zwei

-1.24

ꪻ

-1.22

 dankbar

-1.22

acce

-1.21

หนึ่ง

-1.21

OUBLE

-1.20

וחות

-1.20

 några

-1.16

POSITIVE LOGITS

you

1.96

of

1.86

 that

1.62

 they

1.57

it

1.55

1.53

uyasha

1.52

1.49

 however

1.46

 nawet

1.44

Activations Density 0.029%