INDEX

Explanations

prepositions followed by common words

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

 защото

-1.62

 nerede

-1.41

tzw

-1.37

而不是

-1.35

 biß

-1.34

 sogenannten

-1.32

 antiga

-1.30

鹋

-1.30

 dvě

-1.29

遢

-1.27

POSITIVE LOGITS

 this

1.41

1.32

 includes

1.21

one

1.16

all

1.15

monių

1.14

 enables

1.13

σεων

1.13

orical

1.13

 monast

1.11

Activations Density 0.140%