INDEX

Explanations

lists of items separated by commas and "and"

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

åĨ³

-0.07

éĢģåĪ°äºĨ

-0.07

oyer

-0.07

zio

-0.07

æľīä¸ĢèĤ¡

-0.07

æ±º

-0.07

è¡

-0.07

ä¸ĢèĤ¡

-0.07

ÙħÙĨØª

-0.06

Tec

-0.06

POSITIVE LOGITS

çŃīçŃī

0.07

oth

0.07

early

0.06

 ÑĤÐ¾Ð¿

0.06

sf

0.06

ikli

0.06

ç¥¸

0.06

 strtok

0.06

 ÑģÐºÐ¾ÑĢ

0.06

IBC

0.06

Activations Density 0.044%