INDEX

Explanations

self-defense and protection

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

æıĲåĩº

-0.09

å©ª

-0.08

å°¹

-0.08

åĿĿ

-0.08

ä¼ĹæīĢ

-0.08

çł´äº§

-0.08

æ²¿

-0.08

 Sham

-0.08

 proposed

-0.07

ç§ĺå¯Ĩ

-0.07

POSITIVE LOGITS

éĺ²åį«

0.20

éĺ²å¾¡

0.17

-defense

0.16

 defend

0.15

ä¿Ŀåį«

0.15

æŃ¦è£ħ

0.14

åıįåĩ»

0.14

 Ø§ÙĦØ¯ÙģØ§Ø¹

0.14

 defensive

0.14

 defense

0.14

Activations Density 0.073%