INDEX

Explanations

families and their needs

The neuron fires strongly whenever the word “families” (in its plural form) appears.

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

嫆

-2.50

емые

-2.42

олові

-2.36

售价

-2.31

 použití

-2.16

hões

-2.11

 hábiles

-2.06

⑉

-2.05

 impetus

-2.03

‌هایی

-2.02

POSITIVE LOGITS

2.55

of

2.28

vantaged

2.25

2.23

2.22

 Then

2.20

辚

2.17

2.08

),

2.05

Activations Density 0.006%