INDEX

Explanations

names ending in syllables like Lin, Hong, Que, sam

The neuron strongly activates on proper names of people, especially East Asian personal names.

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

and

-2.09

齣

-1.47

 こんな

-1.34

抬起头

-1.32

 poziom

-1.31

aried

-1.31

 voire

-1.30

あっという間に

-1.30

estimated

-1.30

身邊

-1.30

POSITIVE LOGITS

 also

1.67

</em>

1.63

who

1.59

ůže

1.37

 jotka

1.34

batim

1.34

もある

1.33

鹋

1.32

 voidaan

1.30

 вентиля

1.28

Activations Density 0.010%