INDEX

Explanations

it [descriptor]

The neuron fires on the pronoun “it,” especially when used as an object placeholder (e.g. in “find it …” constructions).

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

棪

-1.93

华

-1.80

isable

-1.76

＊

-1.70

ámide

-1.64

śmy

-1.63

旖

-1.63

 artikke

-1.62

Задачи

-1.61

 joli

-1.61

POSITIVE LOGITS

typeName

1.79

’

1.73

 fueron

1.73

is

1.69

႐

1.66

或者

1.65

or

1.61

鮠

1.59

 will

1.59

1.58

Activations Density 0.007%