INDEX

Explanations

code or data

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

 whole

-0.31

 manifest

-0.27

 odpowied

-0.26

åİĭç¼©

-0.26

 Whole

-0.26

 voiced

-0.25

loh

-0.25

rios

-0.24

 invaded

-0.24

 invite

-0.24

POSITIVE LOGITS

 geography

0.33

çī§

0.29

 geographical

0.29

 territory

0.29

åľ°çĲĨ

0.28

 anatomy

0.27

æ³ķå¾ĭè§Ħå®ļ

0.27

åĨ¶

0.27

åľ°å½¢

0.26

 instrumentation

0.26

Activations Density 0.001%