INDEX

Explanations

numbers and common words

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

AGO

-0.25

QUEST

-0.25

è§Ĩ

-0.24

 Briggs

-0.23

è¯Ĭæ²»

-0.23

åıĸä»£

-0.23

letic

-0.23

 replaced

-0.23

Replace

-0.23

 replacing

-0.23

POSITIVE LOGITS

åħ¶ä»ĸäºº

0.31

 others

0.31

æĻ®éģį

0.30

 universal

0.30

 elsewhere

0.29

åħ¶ä»ĸçİ©å®¶

0.28

åħ¶ä½Ļ

0.28

åĲĮæľŁ

0.27

å¤§å¤ļæķ°äºº

0.27

åħ¬è®¤çļĦ

0.27

Activations Density 0.185%