INDEX

Explanations

grace and adornment

The neuron fires on verbs (and their inflected forms) that mean “to adorn” or “to decorate,” notably “grace” and “adorn.”

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

伐

-1.70

ی

-1.69

zeć

-1.63

ка

-1.61

</h3>

-1.58

 Advertisement

-1.56

 craftsman

-1.55

ᱽ

-1.55

zień

-1.54

ים

-1.52

POSITIVE LOGITS

the

2.13

الدرا

1.95

这

1.88

這

1.85

iting

1.83

المصدر

1.81

枓

1.77

榀

1.75

 potret

1.72

Ито

1.71

Activations Density 0.009%