INDEX

Explanations

as

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

fjspx

-0.72

TagMode

-0.68

 >=",

-0.64

 محفوظة

-0.63

 ivelany

-0.62

MessageTagHelper

-0.59

ⓧ

-0.58

 geldig

-0.55

 تضيفلها

-0.55

adecimal

-0.53

POSITIVE LOGITS

landes

0.54

 okazji

0.51

hpp

0.50

../../../

0.47

 gewinnt

0.45

 Zuge

0.45

chir

0.44

ிறது

0.44

hol

0.44

 Negara

0.44

Activations Density 0.007%