INDEX

Explanations

names of people and places

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 ÐĲÐ»ÐµÐºÑģÐ°Ð½

-0.26

 ÐŁÐ¾ÑĩÐµÐ¼

-0.25

à¹Ģà¸Ĺà¸Ħà¹Ĥà¸Ļ

-0.24

 Ð¿ÑĢÐ¸ÑıÑĤÐ½

-0.22

à¸Ħà¸°à¹ģà¸Ļ

-0.22

DataTask

-0.20

à¹Ĥà¸£à¸ĩà¸ŀà¸¢à¸²

-0.20

à¸Ħà¸ĵà¸°à¸ģà¸£à¸£à¸¡

-0.20

 à¸¡à¸ģà¸£

-0.20

ropp

-0.20

POSITIVE LOGITS

ãĢįãĢĮ

0.21

ç¶ļãģĦãģ¦

0.20

 Burns

0.19

äººå®¶

0.18

åı¦

0.17

ä¸¤åıª

0.17

ç²¾èĭ±

0.17

åģıåĲĳ

0.17

æĬ¡

0.17

dump

0.17

Activations Density 0.536%

No Known Activations