INDEX

Explanations

bracketed phrases and proper nouns

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

 бампер

-1.39

 silnika

-1.31

ję

-1.29

pantalon

-1.25

==""){

-1.25

 escond

-1.19

 vicini

-1.19

 łyż

-1.19

get

-1.17

banyak

-1.16

POSITIVE LOGITS

一方面

1.37

 תמיד

1.24

 zarówno

1.23

DEA

1.22

 државе

1.20

 塗装

1.18

 preso

1.16

APOLIS

1.15

偶尔

1.15

both

1.15

Activations Density 0.002%