INDEX

Explanations

Start of a text

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

çĶľ

-0.27

 pitch

-0.27

æ³ķå¾ĭæ³ķè§Ħ

-0.25

Åģ

-0.25

æĸĹ

-0.25

Pitch

-0.25

åĭ¾

-0.25

çĻ»

-0.25

 Privacy

-0.25

javax

-0.24

POSITIVE LOGITS

ends

0.26

æģ½

0.25

æ¶Īè²»èĢħ

0.25

setType

0.25

éĢļçķħ

0.24

ders

0.24

die

0.24

çĻį

0.24

éĤ¨

0.24

 retailers

0.23

Activations Density 0.001%