INDEX

Explanations

software licenses

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

in

-0.36

åľ¨åĨħçļĦ

-0.32

in

-0.31

inan

-0.30

 dazu

-0.29

æīĵçĿĢ

-0.27

akah

-0.27

XHR

-0.27

inja

-0.26

iang

-0.25

POSITIVE LOGITS

æľº

0.29

éĴ¤

0.27

 resident

0.27

adel

0.26

nes

0.25

/pages

0.25

ogan

0.24

esson

0.24

ç»ĵ

0.24

éĹ®

0.24

Activations Density 0.897%