INDEX

Explanations

Proper nouns and specific terms

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

CALL

-0.83

/></

-0.82

 clar

-0.80

entification

-0.79

Ngb

-0.79

ütze

-0.78

 disque

-0.78

@}

-0.77

 sué

-0.77

 Dép

-0.75

POSITIVE LOGITS

{-

1.29

--

1.05

--

0.91

 instance

0.84

"--

0.82

 ziren

0.81

 where

0.81

 Sortiment

0.81

Inhalte

0.80

Theatre

0.79

Activations Density 0.064%