INDEX

Explanations

data structures array

The neuron strongly activates on mentions of arrays (including “array” and its equivalents like the Russian “массив”), especially in contexts involving large or “massive” data.

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

 حتی

-1.84

 indicato

-1.71

بغ

-1.55

』

-1.55

 grazia

-1.53

谙

-1.52

席

-1.52

-1.50

hoka

-1.49

Even

-1.45

POSITIVE LOGITS

</strong>

1.68

</h3>

1.68

of

1.65

シェフ

1.57

1.55

our

1.55

ネイル

1.51

 RECOMM

1.51

 firmar

1.50

色が

1.49

Activations Density 0.041%