INDEX

Explanations

listing items with "and"

conjunctions and lists

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

Ბ

0.26

HomeScreen

0.24

ูนย์

0.24

<start_of_turn>

0.24

嗶

0.23

Cooldown

0.22

鲈

0.22

AlH

0.22

मलव

0.22

<unused52>

0.22

POSITIVE LOGITS

and

0.58

and

0.48

etc

0.46

và

0.44

и

0.39

 και

0.39

 ಮತ್ತು

0.38

 आणि

0.38

و

0.38

 और

0.37

Activations Density 0.475%