INDEX

Explanations

thinking about

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

 Claims

-0.22

Ð¿ÑĢÐ°ÐºÑĤÐ¸Ðº

-0.21

Ð¿ÑĢÐ¾ÑĦÐµÑģÑģÐ¸Ð¾Ð½Ð°Ð»

-0.21

ä¸įæĪĲ

-0.20

 mayor

-0.20

Ð¼ÐµÐ´Ð¸ÑĨÐ¸Ð½ÑģÐº

-0.20

Ð¸Ð½Ð´Ð¸Ð²Ð¸Ð´

-0.20

Ð¸Ð½Ð´Ð¸Ð²

-0.20

åĨħå¿ĥçļĦ

-0.19

æ·ĺ

-0.19

POSITIVE LOGITS

 about

0.31

cape

0.29

æĥ³çĿĢ

0.28

å¿ĸ

0.26

ä¸ĩåįĥ

0.26

-thinking

0.25

èĻĳ

0.25

 aloud

0.25

ahead

0.25

 thoughts

0.25

Activations Density 0.064%