INDEX

Explanations

actions taken from links

This neuron fires on direct audience-address language—especially first-person plural (“we,” “our”) and second-person (“you,” “your”) pronouns used in calls to action, disclosures, and engagement phrases.

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

in

-1.32

 wants

-1.22

（

-1.22

on

-1.21

vimeo

-1.20

トヨタ

-1.20

at

-1.20

 purposefully

-1.19

for

-1.19

or

-1.18

POSITIVE LOGITS

 dando

1.35

苙

1.30

 …..

1.30

[],

1.27

 búsquedas

1.26

 .......

1.25

れません

1.24

 nhung

1.22

 '-':

1.21

೦

1.21

Activations Density 0.091%