INDEX

Explanations

This neuron detects direct second-person address or reader-oriented discourse markers—phrases that call out “you” (e.g. “as you might know,” “for those of you,” “there’s no denying the fact that you…,” etc.).

confessions and admissions

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

»;

-1.30

 kabát

-1.27

licopter

-1.25

łoni

-1.23

łania

-1.19

cześ

-1.19

БУ

-1.19

مك

-1.16

oooooooo

-1.16

 maleta

-1.16

POSITIVE LOGITS

of

1.43

外部連結

1.14

—

1.11

1.09

楽しめます

1.09

 kişilik

1.07

随之

1.05

 manoeu

1.05

—

1.04

Activations Density 0.170%