INDEX

Explanations

attend / participate

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

å©ļå§»

-0.09

ç»¼åĲĪåĪ©çĶ¨

-0.09

æ¯ıæĹ¥

-0.08

åľ¨çĶŁæ´»ä¸Ń

-0.08

æĿĢæŃ»

-0.08

è¾¾æĪĲ

-0.08

opo

-0.08

conversation

-0.08

çī¹è´¨

-0.08

è¾Ľ

-0.08

POSITIVE LOGITS

åıĤåĬł

0.25

åıĤä¸İ

0.23

åıĤåĬłäºĨ

0.20

 participation

0.19

åıĥåĬł

0.19

 Participation

0.18

åıĥèĪĩ

0.17

 participated

0.17

 participate

0.16

åĩºå¸Ń

0.16

Activations Density 0.135%