INDEX

Explanations

categories and specific items

The neuron is activated by tokens that represent numeric quantities (digits and measures).

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

 τὴν

-1.55

 nuestra

-1.50

we

-1.47

ἡ

-1.40

сибир

-1.38

 ἐκ

-1.37

 τὰ

-1.36

ophyta

-1.33

 النبات

-1.30

 chrétienne

-1.28

POSITIVE LOGITS

 those

1.52

 السياس

1.40

ּוֹ

1.38

 kwenye

1.30

他們

1.29

ֵּ

1.27

ῤ

1.26

fibra

1.24

他们

1.21

ִּ

1.21

Activations Density 0.016%