INDEX

Explanations

referring to her or she

The neuron strongly responds to the very first content token in each new text segment (i.e. the first word right after a segment boundary).

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

-2.42

,'

-1.95

 tardes

-1.79

谞

-1.77

_'

-1.77

 mouseY

-1.76

ⓘ

-1.74

 intuitive

-1.72

</em>

-1.69

});*/

-1.68

POSITIVE LOGITS

 ‘‘

1.70

銲

1.70

 faceva

1.66

푉

1.63

 ۱۲

1.63

 depicted

1.62

 voleva

1.61

 început

1.60

 riusc

1.60

 poteva

1.60

Activations Density 0.016%