INDEX

Explanations

scientific and technical terms

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

 barrel

-0.25

/from

-0.24

inite

-0.22

Ĥ¬

-0.22

seg

-0.22

 forefront

-0.22

à¸ļà¸²

-0.21

:↵

-0.21

 entirety

-0.21

POSITIVE LOGITS

çłĶç©¶è¡¨æĺİ

0.43

åıĭä»¬

0.35

è°ĥæŁ¥æĺ¾ç¤º

0.35

è¶Ĭæĺ¯

0.29

æķ°æį®æĺ¾ç¤º

0.29

äººå£«

0.29

åŃ¦å®¶

0.29

 researchers

0.27

èµĦæĸĻæĺ¾ç¤º

0.27

 experts

0.26

Activations Density 0.365%