INDEX

Explanations

even if

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

jian

-0.07

å®ŀéĻħæİ§åĪ¶æĿĥ

-0.07

Ð³ÐµÐ»ÑĮ

-0.07

Ð¾Ð±Ðµ

-0.06

åĵªäºĽéĹ®é¢ĺ

-0.06

ÑģÑĤÐºÐ°

-0.06

atora

-0.06

AINER

-0.06

neau

-0.06

è´Łè´£åĪ¶

-0.06

POSITIVE LOGITS

dur

0.06

0.05

 vá»¥

0.05

 devil

0.05

å°Ĩ

0.05

éģį

0.05

uesa

0.05

åľ°æĸ¹

0.05

 regard

0.04

ìļ©

0.04

Activations Density 0.007%