INDEX

Explanations

descriptions of appearance and function

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

or

-2.59

to

-2.14

but

-2.06

in

-2.03

 however

-1.78

However

-1.71

 jedoch

-1.55

of

-1.52

as

-1.50

but

-1.45

POSITIVE LOGITS

hmig

1.55

謦

1.49

༷

1.45

ڤ

1.40

reportWebVitals

1.39

äsenti

1.36

夘

1.36

 bobina

1.33

 dagar

1.33

 dinam

1.33

Activations Density 0.038%