INDEX

Explanations

seeing, experiencing, or trying something

The neuron fires on direct second‐person address—words like “you,” “your,” and related reader‐directed calls to action.

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

you

-1.38

 there

-1.34

 atriz

-1.28

oprote

-1.24

vább

-1.16

 proy

-1.15

amide

-1.14

-1.13

 made

-1.12

 анализа

-1.12

POSITIVE LOGITS

 দেখ

1.71

see

1.41

 확인

1.38

oldown

1.20

 understand

1.19

 know

1.19

🏟

1.18

ichtigen

1.15

événement

1.14

These

1.13

Activations Density 0.061%