INDEX

Explanations

By using, is provided, able to

concepts and states

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

Categorie

0.49

 दादा

0.48

Expanded

0.47

App

0.46

Reed

0.46

CCH

0.46

 بخش

0.45

 Серге

0.45

APPA

0.45

Form

0.45

POSITIVE LOGITS

珑

0.46

 doable

0.45

قتل

0.44

 tứ

0.43

 jatuh

0.43

 sacrificing

0.42

 ショルダーバッグ

0.42

 leukemia

0.42

 curing

0.41

 candy

0.41

Activations Density 0.001%