INDEX

Explanations

most prevalent

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

ä»¥èī²

-0.21

è°Ģ

-0.19

ANGED

-0.19

Called

-0.18

 Stamina

-0.18

ä½ľåĩºäºĨ

-0.18

 rules

-0.17

 sustain

-0.17

æ¸ħæĸ°

-0.17

 ÑĤÐ°ÐºÐ¸Ñħ

-0.17

POSITIVE LOGITS

ä¹ĭä¸Ģ

0.35

åİŁåĽłä¹ĭä¸Ģ

0.23

åĽłä¸ºå®ĥ

0.22

çļĦä¸Ģéĥ¨åĪĨ

0.21

 Ã¼berhaupt

0.19

ä¸»æĮģ

0.19

[#

0.19

åħ¶æ¬¡æĺ¯

0.19

çļĦèº«ä»½

0.19

åĽłä¸ºä»ĸä»¬

0.19

Activations Density 0.284%