INDEX

Explanations

specific location or item

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

oyo

-0.07

éŀ

-0.07

inv

-0.07

å¾®ä¿¡æī«çłģ

-0.07

lander

-0.07

 ÙĪÙĥØ§ÙĦØ©

-0.07

åıĪæĿ¥äºĨ

-0.07

çľĭä¸įåĩº

-0.06

.allowed

-0.06

è¢«æµıè§Ī

-0.06

POSITIVE LOGITS

 exact

0.11

exact

0.11

 location

0.10

 elusive

0.10

 Exact

0.10

åĩĨç¡®çļĦ

0.10

ç²¾åĩĨ

0.10

Exact

0.10

æī¾åĪ°

0.09

 corresponding

0.09

Activations Density 0.314%