INDEX

Explanations

scientific/medical contexts

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

ROP

-0.61

idist

-0.59

ṇ

-0.55

 autorytatywna

-0.53

indale

-0.50

isser

-0.49

tilled

-0.48

liffe

-0.48

abestanden

-0.46

walde

-0.46

POSITIVE LOGITS

 kaynağından

0.63

[],

0.59

 Administrativna

0.57

 avoidance

0.56

appé

0.56

Yikes

0.55

pantalones

0.55

ggiato

0.55

cluso

0.54

 Spart

0.54

Activations Density 0.035%