INDEX

Explanations

why, how, whether, or

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

了一会儿

-1.45

 добраться

-1.39

rijving

-1.36

anterie

-1.30

といった

-1.27

}:

-1.26

ographe

-1.26

to

-1.26

ppelin

-1.24

are

-1.22

POSITIVE LOGITS

it

1.53

 vijf

1.52

 natuurlijk

1.52

 zacht

1.49

why

1.43

or

1.41

↵

1.39

how

1.36

 whether

1.33

 mojej

1.30

Activations Density 0.324%