INDEX

Explanations

abbreviations and specific entities

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 Erectile

-0.28

 paddingHorizontal

-0.21

={"/

-0.21

çľĭäºĨä¸Ģçľ¼

-0.21

æİĴéª¨

-0.20

ALLERY

-0.20

ereotype

-0.19

äº¤æĺĵå¹³åı°

-0.19

æĿ¥çľĭçľĭåĲ§

-0.19

ÑĥÐ½Ð¸ÑĨÐ¸Ð¿

-0.18

POSITIVE LOGITS

":"'

0.22

ÑĦÐµÑģÑģÐ¸Ð¾Ð½Ð°

0.22

æĤĴ

0.21

edom

0.20

.Dial

0.20

aggio

0.20

"crypto

0.20

isphere

0.19

ilder

0.19

bey

0.19

Activations Density 0.001%

No Known Activations