INDEX

Explanations

causes, means, indicates, shows

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

 there

-1.55

 Novembre

-1.37

Это

-1.30

sz

-1.20

--;

-1.18

 communis

-1.16

 Bucure

-1.16

 Seigneur

-1.16

 eser

-1.14

 while

-1.13

POSITIVE LOGITS

 trzech

1.80

1.77

1.48

 duży

1.43

 burbujas

1.40

 jakiś

1.39

digarh

1.39

him

1.38

 ouvido

1.38

mitsub

1.38

Activations Density 0.144%