INDEX

Explanations

first page or results

on a/the line

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

ంక

0.43

ルト

0.42

Fr

0.41

OR

0.38

 annular

0.38

 incorporation

0.38

 flannel

0.37

 pipe

0.36

 bookmarks

0.36

Vn

0.36

POSITIVE LOGITS

 birinci

0.47

 langit

0.46

 değil

0.45

 jedną

0.45

 आघाडी

0.45

 oberen

0.45

yaxis

0.44

 fejl

0.44

 deutlich

0.44

 होईल

0.44

Activations Density 0.001%