INDEX

Explanations

"seems messy" or "instead"

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

æĹ¢åı¯ä»¥

-0.22

Aus

-0.21

 laps

-0.19

åīįæīĢæľªæľī

-0.19

lap

-0.19

ä¸ĭç§°

-0.18

ocol

-0.18

 Rach

-0.18

LAP

-0.18

ä¸įä»ħåı¯ä»¥

-0.17

POSITIVE LOGITS

å¤ªä¹ħ

0.27

å¤ª

0.26

å¤ªé«ĺ

0.25

ä¸įè¯´

0.24

ä¹Łä¸įæĥ³

0.23

åģĥ

0.21

ç¥ŀä»Ļ

0.21

too

0.21

éĻ¤éĿŀ

0.20

ä¸įå®ľ

0.20

Activations Density 0.220%