INDEX

Explanations

manipulate and edit

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

à§Ī

-0.08

åİ

-0.07

åĩĨåĪĻ

-0.07

emel

-0.07

ãģĹãģĭãĤĤ

-0.07

_scal

-0.07

å¹¶å·²

-0.07

uerte

-0.07

 certes

-0.07

ukes

-0.07

POSITIVE LOGITS

å¿ħè¦ģæĹ¶

0.12

ä»¥éĺ²

0.12

æĪĸçĽ´æİ¥

0.11

ä»¥ç¡®ä¿Ŀ

0.11

 sometimes

0.10

ä»¥éģ¿åħį

0.10

å¿ħè¦ģçļĦ

0.10

 ÑĩÑĤÐ¾Ð±Ñĭ

0.10

æĪĸèĢħçĽ´æİ¥

0.10

å¦ĤæŀľéľĢè¦ģ

0.10

Activations Density 0.107%