INDEX

Explanations

property, setting, finding

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

 hitta

-1.58

and

-1.42

 inhalt

-1.35

engono

-1.33

 hjälpa

-1.30

 tyme

-1.28

ayage

-1.23

shaft

-1.21

êter

-1.20

 作り方

-1.19

POSITIVE LOGITS

 that

1.79

 then

1.73

or

1.52

们

1.48

 than

1.48

犼

1.43

юрпри

1.34

if

1.34

 something

1.34

多く

1.30

Activations Density 0.355%