INDEX

Explanations

supplement or complement

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

_exc

-0.10

exc

-0.09

è¤°

-0.09

çīµ

-0.09

çīµæĮĤ

-0.09

umed

-0.09

çļĭ

-0.08

åĢ¾æĸľ

-0.08

åįıä¼ļä¼ļéķ¿

-0.08

åĢ¾åĲĳ

-0.08

POSITIVE LOGITS

äºĴè¡¥

0.16

 complement

0.14

 complimentary

0.13

 complementary

0.13

è¡¥åħħ

0.12

 existing

0.12

 Existing

0.12

çİ°æľīçļĦ

0.12

 compliment

0.11

çİ°æľī

0.11

Activations Density 0.070%