INDEX

Explanations

design including designing sets

users choose or vote

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

LDL

0.47

න

0.46

 Али

0.46

𝙻

0.46

 மனைவி

0.46

 großes

0.46

 علي

0.45

ley

0.45

 али

0.45

 അധിക

0.45

POSITIVE LOGITS

 stamped

0.53

在

0.49

 notebook

0.48

 metall

0.48

 tertarik

0.46

bol

0.45

 smer

0.45

 chronograph

0.44

 prest

0.44

但我

0.44

Activations Density 0.001%