INDEX

Explanations

math problems

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

äº§

-0.33

 uniquely

-0.26

 enormously

-0.26

 davon

-0.25

omed

-0.25

æĬ¹

-0.25

 bump

-0.24

_REUSE

-0.24

-rated

-0.24

 Savage

-0.24

POSITIVE LOGITS

å¸¸

0.28

åħ¨éĿ¢æİ¨è¿Ľ

0.27

é¢ĳ

0.27

esp

0.26

 inkl

0.26

PTY

0.25

åºĶåĬĽ

0.25

eva

0.25

edy

0.25

alic

0.24

Activations Density 0.102%