INDEX

Explanations

only

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

 already

-0.23

almost

-0.23

 sigu

-0.22

fkk

-0.22

ãģ¾ãģļ

-0.22

already

-0.22

 {?>↵

-0.22

 almost

-0.21

ä¸įæŃ¢

-0.20

DMETHOD

-0.20

POSITIVE LOGITS

åįķçº¯çļĦ

0.29

èĢĮå·²

0.28

ifi

0.27

çŁŃçŁŃ

0.26

:".$

0.25

ONE

0.25

å°ĳæķ°

0.24

äºĽè®¸

0.24

*****/↵

0.24

chedulers

0.24

Activations Density 1.087%