INDEX

Explanations

questions starting with what

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

 also

-1.78

 также

-1.70

 และ

-1.54

 cauſe

-1.41

 également

-1.38

 própri

-1.36

 також

-1.35

也非常

-1.34

 durer

-1.33

 dito

-1.29

POSITIVE LOGITS

 four

1.47

 those

1.44

jenigen

1.33

Those

1.30

 that

1.27

 five

1.25

selben

1.23

Manfaat

1.22

LETIN

1.22

 entweder

1.22

Activations Density 0.015%