INDEX

Explanations

clarity on, believe or, news the

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

æĪĳå¯¹

-0.24

ä½łä¼ļåıĳçİ°

-0.21

lement

-0.21

ÑĶ

-0.21

ãķ

-0.20

éĴ¯

-0.20

éĽ©

-0.20

<_

-0.20

éĺ¿å§¨

-0.20

å²«

-0.20

POSITIVE LOGITS

ä¸ºåħ¶

0.36

 themselves

0.30

åľ¨åħ¶

0.29

ä»ĸä»¬çļĦ

0.28

ç»Ļä»ĸä»¬

0.27

å¯¹ä»ĸä»¬

0.26

ä½Ĩä»ĸä»¬

0.25

è®©ä»ĸä»¬

0.23

angs

0.23

acks

0.23

Activations Density 0.236%