INDEX

Explanations

archivi properties

This neuron strongly activates on long Russian nouns (Cyrillic words denoting concepts, like “предложения,” “закон,” or “сохранения”).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

of

-3.72

you

-2.38

we

-1.97

if

-1.93

it

-1.92

 there

-1.75

 even

-1.63

-1.58

min

-1.55

 jonka

-1.55

POSITIVE LOGITS

intendo

1.75

</em>

1.66

basti

1.61

насе

1.58

vlád

1.58

</strong>

1.53

archivi

1.53

jakarta

1.51

贺

1.50

tapis

1.49

Activations Density 0.377%