INDEX

Explanations

words or phrases ending a sentence

why

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

ol

0.57

 Moses

0.47

ali

0.46

unders

0.46

ajouter

0.45

oki

0.45

urt

0.45

 Bones

0.45

ad

0.44

POSITIVE LOGITS

 kebut

0.47

vay

0.47

樎

0.43

 Technologien

0.43

 geométricas

0.42

 matin

0.42

碐

0.42

rischen

0.41

ിലി

0.40

अय

0.40

Activations Density 0.001%