INDEX

Explanations

true or false outcomes

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

Споља

-0.97

Podob

-0.93

Современ

-0.88

̔

-0.87

ఉ

-0.87

مدیر

-0.86

billed

-0.84

ఆ

-0.84

Без

-0.83

آیا

-0.83

POSITIVE LOGITS

 false

1.44

 null

1.31

 returns

1.22

or

1.14

 NULL

1.08

 zero

1.08

 interessieren

1.00

 else

0.97

 либо

0.96

 nullptr

0.95

Activations Density 0.021%