INDEX

Explanations

treatment or claims

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

imson

-0.08

ä¾£

-0.08

æ²¹æ³µ

-0.07

é²¤é±¼

-0.07

è¯¾æĶ¹

-0.07

agar

-0.07

rans

-0.07

oley

-0.07

 fled

-0.06

Ã´i

-0.06

POSITIVE LOGITS

trace

0.07

ç¥ŀå±±

0.07

å¹´åĮĸ

0.07

.Interop

0.07

 trace

0.07

æĭ¿æįı

0.06

Ð¸ÑģÑı

0.06

å¤§æĪĲ

0.06

å°±åľ¨

0.06

é©¾é©¶æľºåĬ¨è½¦

0.06

Activations Density 0.004%