INDEX

Explanations

before, preceding, previous, precedent

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

èĭ±æł¼

-0.11

ä¸Ģæĸ¹

-0.09

ÐŁÐ¾Ð´ÑĢÐ¾Ð±Ð½ÐµÐµ

-0.09

çĿ¦

-0.09

ä¸Ģç¯ĩ

-0.09

è¾«

-0.09

 eget

-0.09

 ÑģÐ²Ð¾ÐµÐ¼

-0.08

åıµ

-0.08

å¸¸è¯´

-0.08

POSITIVE LOGITS

åīįè¾Ī

0.16

åīįéĿ¢

0.16

preced

0.16

 preced

0.15

 preceding

0.15

åħĪè¡Į

0.14

åľ¨æŃ¤ä¹ĭåīį

0.12

 precedent

0.12

èº«åĲİ

0.11

åĲİéĿ¢

0.11

Activations Density 0.058%

No Known Activations