INDEX

Explanations

both, all leading to plural items

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

noop

-0.08

èļĬ

-0.07

å¯¹å®ĥ

-0.07

:class

-0.07

lickr

-0.07

bo

-0.07

abo

-0.07

æ³¼

-0.07

à¹Ģà¸¥

-0.07

stra

-0.07

POSITIVE LOGITS

è¿ĻäºĽéĥ½æĺ¯

0.08

è¿ĻäºĽéĥ½

0.08

)||(

0.08

 together

0.07

dov

0.07

 zusammen

0.07

åĳ¨åĪ°çļĦ

0.07

åĿĩä¸º

0.07

 latter

0.06

åĵªä¸ªå¥½

0.06

Activations Density 0.062%