INDEX

Explanations

access to destinations

The neuron detects mentions of proximity or location—tokens that express distance, directions, or nearby transit/stops (e.g. “to,” “near,” “close,” “from,” specific lines or squares).

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

 hiasan

-1.43

 that

-1.39

during

-1.34

によって

-1.27

tocks

-1.22

 outubro

-1.20

rogens

-1.20

 UITextField

-1.19

 março

-1.18

wdriver

-1.17

POSITIVE LOGITS

 both

2.16

 everything

1.60

all

1.59

ทั้ง

1.59

Both

1.45

 great

1.20

 everyone

1.16

 several

1.16

所有的

1.16

 various

1.15

Activations Density 0.010%