INDEX

Explanations

States and countries names

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

 when

-2.16

 this

-1.94

 while

-1.86

for

-1.79

 with

-1.58

طط

-1.52

to

-1.50

 during

-1.49

 from

-1.46

 mulia

-1.39

POSITIVE LOGITS

 eerste

1.84

 моей

1.45

 első

1.44

 erzä

1.35

 отец

1.33

穴

1.33

 первое

1.30

 Verkäufer

1.28

 моих

1.28

my

1.27

Activations Density 0.125%