INDEX

Explanations

Varied text sources

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

 passports

-0.27

duct

-0.25

 campuses

-0.25

ETIME

-0.25

åº¹

-0.24

åŁł

-0.24

enticator

-0.24

å®ħ

-0.24

éļĻ

-0.24

dress

-0.24

POSITIVE LOGITS

OLLOW

0.27

 fork

0.25

ä¹ĺ

0.25

itan

0.25

inner

0.25

å¹³æĿ¿

0.25

æŃ¥ä¼Ĳ

0.24

æ³ī

0.24

ç¾Ĭ

0.24

linked

0.23

Activations Density 0.022%