INDEX

Explanations

number followed by min

sample descriptions

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

 एहसास

0.49

,’

0.48

⟺

0.48

 democracy

0.47

democracy

0.46

 Democracy

0.45

 Conquer

0.44

<code>

0.43

 afore

0.42

लेस

0.42

POSITIVE LOGITS

ürdig

0.50

 Gruppe

0.49

 Bericht

0.48

 సాహిత్య

0.47

ILITIES

0.47

CHREIB

0.46

ITIONAL

0.46

ពិស

0.46

IEN

0.46

摖

0.45

Activations Density 0.000%