INDEX

Explanations

numbers followed by ordinals

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

ðŁĺĽ

-0.20

ÑĤÐ°Ð½

-0.20

ðŁĺ¦

-0.20

Ð·Ð¾ÑĢ

-0.18

uhe

-0.18

::::::

-0.18

 âĺĨ

-0.18

ÐºÐ°Ð½

-0.18

ðŁĺĲ

-0.17

olle

-0.17

POSITIVE LOGITS

ï¸ı

0.38

åĦ¿

0.33

à¹Į

0.28

åħĴ

0.27

Ùĳ

0.27

åŃĲéĩĮ

0.26

0.25

es

0.25

à¹Ĩ

0.24

åħ®

0.24

Activations Density 0.069%