INDEX

Explanations

unique and not the same

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

pose

-0.26

å»ºè®®

-0.25

(crate

-0.25

æľīå¿ħè¦ģ

-0.24

èŁ¾

-0.23

rac

-0.23

 contr

-0.22

ä¸įåºĶè¯¥

-0.22

se

-0.22

 Weak

-0.22

POSITIVE LOGITS

tÃŃ

0.26

ä¸ĢåĽŀäºĭ

0.26

 Rockefeller

0.25

 Bentley

0.23

åĽŀäºĭ

0.23

imson

0.23

idth

0.23

Nor

0.22

 hostages

0.22

imi

0.22

Activations Density 0.017%