INDEX

Explanations

No Explanations Found

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

 sürd

-0.45

 ***!

-0.45

 Bibliothèque

-0.44

roek

-0.44

 EnglishChoose

-0.43

__*/

-0.42

 situe

-0.42

rumah

-0.42

 تانيه

-0.41

묻

-0.41

POSITIVE LOGITS

 spawn

0.68

hob

0.66

 crimp

0.63

IndentedString

0.63

ergies

0.63

gum

0.62

verifyException

0.61

NOPQRST

0.60

 gawas

0.60

up

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.