INDEX

Explanations

names of people and places

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

 блюдо

-1.42

 zucca

-1.26

 yani

-1.23

 religieuses

-1.22

PTSD

-1.20

 chré

-1.16

 сайті

-1.16

ÁN

-1.14

 牛肉

-1.13

留意

-1.13

POSITIVE LOGITS

ּוֹ

1.94

 during

1.46

 السياس

1.45

our

1.39

 teilweise

1.35

in

1.33

ְּ

1.31

ִּ

1.30

 кроссовки

1.27

 when

1.27

Activations Density 0.011%