INDEX

Explanations

numbers, units, and measurements

The neuron responds to document‐structure and formatting markers (section tags, heading underlines, soft hyphens, special spacing, etc.).

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

it

-2.14

 these

-1.82

 this

-1.70

 those

-1.57

 some

-1.54

：……

-1.53

our

-1.52

-1.51

the

-1.48

が必要です

-1.47

POSITIVE LOGITS

orative

1.79

yled

1.55

omsday

1.49

lised

1.48

letoe

1.45

thodoxy

1.45

 パーカー

1.44

ibly

1.44

istas

1.44

|[]

1.42

Activations Density 0.330%