INDEX

Explanations

project root or include

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

 FALSE

-0.73

cheln

-0.73

toolbar

-0.72

 BOOLEAN

-0.71

 Impedance

-0.71

 cubos

-0.70

 Carbide

-0.68

quellen

-0.68

 RULE

-0.67

lino

-0.67

POSITIVE LOGITS

ր

0.66

 reunión

0.65

 mengingatkan

0.65

ｭ

0.64

inov

0.64

 dispatch

0.63

慣れて

0.63

ждению

0.63

 Veteran

0.62

appspot

0.60

Activations Density 0.086%