INDEX

Explanations

studies involving adults

The neuron is primarily detecting numeric tokens—counts, percentages, years, and other statistics—within scientific or research text.

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

ׇ

-1.83

ὴν

-1.75

 gelas

-1.66

 chré

-1.64

entibus

-1.63

 disiplin

-1.62

 eccl

-1.56

 aktivi

-1.55

 bicchiere

-1.54

izamos

-1.53

POSITIVE LOGITS

ּוֹ

2.23

ֶּ

2.16

 бампер

2.09

ִּ

1.91

ְּ

1.68

וּ

1.66

־ה

1.54

ַּ

1.54

ֵּ

1.52

ָּ

1.50

Activations Density 0.827%