INDEX

Explanations

dimensions and components of structures

This neuron fires on mentions of specific built‐structure or vehicle model names and their size/feature descriptors (e.g. “Tiny House,” “Prototype,” “5th Wheel,” “modular construction”).

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

as

-1.53

in

-1.53

 even

-1.45

 particular

-1.23

 noticeable

-1.16

જ

-1.16

 ПОД

-1.16

 first

-1.16

 Particular

-1.14

 during

-1.13

POSITIVE LOGITS

 dedo

1.31

 entièrement

1.30

 voulais

1.29

 complètement

1.25

︱

1.23

 våra

1.21

 sommeil

1.21

espejo

1.20

 Этот

1.18

bakter

1.17

Activations Density 0.115%