INDEX

Explanations

journal citations and page numbers

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

çļĦæ¯įäº²

-0.24

çĭİ

-0.24

é¦ĸæ¬¡

-0.24

(wp

-0.23

è¯Ģ

-0.23

åĲ¯åĬ¨ä»ªå¼ı

-0.23

åĴı

-0.22

çļĦçĪ¶äº²

-0.22

æĹ©çĤ¹

-0.22

æģ¯

-0.21

POSITIVE LOGITS

 Bulletin

0.22

Bulletin

0.22

{%

0.22

 motivating

0.22

 conscient

0.21

çŁŃæľŁåĨħ

0.21

riters

0.21

 Notices

0.20

æ¯ıä¸Ģä¸ª

0.20

 occupied

0.20

Activations Density 0.044%