INDEX

Explanations

greetings and conversational starters

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

ä½ĵåĪ¶

-0.24

è¯łéĩĬ

-0.23

è¿ĺåĮħæĭ¬

-0.22

 Opport

-0.22

ÐĵÐ»Ð°Ð²Ð½Ð°Ñı

-0.22

æıŃéľ²

-0.22

(pg

-0.22

å¹¿æ³ĽçļĦ

-0.22

æīĢå¸¦æĿ¥çļĦ

-0.22

 organised

-0.21

POSITIVE LOGITS

your

0.28

 sorry

0.27

æĬ±æŃī

0.27

try

0.26

Ĥ¨

0.25

åĹ¨

0.25

hey

0.24

ç¹ģå¿Ļ

0.24

 your

0.24

å¯¹ä¸įèµ·

0.24

Activations Density 0.015%