INDEX

Explanations

how things are seen or perceived

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

 Both

-1.63

is

-1.55

该

-1.48

</b>

-1.44

Both

-1.39

 Instead

-1.34

Either

-1.32

 دندان

-1.31

тием

-1.27

to

-1.25

POSITIVE LOGITS

entown

1.50

of

1.42

gambarkan

1.31

 then

1.31

῎

1.30

jących

1.26

 trastornos

1.25

 dinosaurio

1.25

YNAMICS

1.24

みを

1.19

Activations Density 0.017%