INDEX

Explanations

technical terms and keywords

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

and

-2.17

for

-1.63

or

-1.58

who

-1.58

-1.48

of

-1.45

 which

-1.40

to

-1.39

 επίσης

-1.29

 šir

-1.25

POSITIVE LOGITS

 ocultar

1.31

 nettoyer

1.27

 vuol

1.25

 oreille

1.25

 réinitial

1.24

Achter

1.23

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.23

 monaster

1.22

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.22

Hersteller

1.21

Activations Density 0.053%