INDEX

Explanations

in or by

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

.Escape

-0.24

rary

-0.23

aset

-0.23

 bitten

-0.22

 Rape

-0.22

amar

-0.22

neh

-0.22

é¦Į

-0.21

nehmer

-0.21

}elseif

-0.21

POSITIVE LOGITS

çĲĨè§£åĴĮ

0.29

é£İéĩĩ

0.26

åŃķ

0.26

è¿ĻçĤ¹

0.26

æŃ¤åŁºç¡Ģä¸Ĭ

0.25

èĩªä¿¡

0.25

æĹ¶éĹ´åĴĮ

0.24

è¡Įèµ°

0.24

ä¸įä½ı

0.24

å¾Ĺä½ı

0.24

Activations Density 0.017%