INDEX

Explanations

return statements in code

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

ixes

-1.31

 sehr

-1.30

تصویر

-1.26

zation

-1.24

all

-1.23

be

-1.16

مشخصات

-1.16

 each

-1.15

 assess

-1.15

 fibras

-1.12

POSITIVE LOGITS

 VICE

1.30

膑

1.27

 Сколько

1.26



1.23

CHILDREN

1.23

PINES

1.23

都已经

1.23

 教程

1.20

 Años

1.20

GOING

1.18

Activations Density 0.001%