INDEX

Explanations

The neuron strongly activates on words referring to living creatures or organisms.

like a

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

ּוֹ

-1.67

ִּ

-1.28

 бампер

-1.22

ֶּ

-1.19

ַּ

-1.06

 κοιν

-1.05

観察

-1.03

cluded

-1.03

 השאלה

-1.02

 montagnes

-1.00

POSITIVE LOGITS

ׇ

1.47

ֳ

1.21

or

1.09

 støtte

1.07

ֺ

1.00

 crude

0.99

応え

0.98

<0xA3>

0.97

 særlig

0.96

 våre

0.94

Activations Density 0.018%