INDEX

Explanations

before or after

The neuron strongly activates on common prepositions—especially “in,” “before,” and similar temporal/location markers.

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

-1.97

 Wanneer

-1.59

 encantador

-1.50

匱

-1.48

 Когда

-1.47

 acondicionado

-1.41

훑

-1.41



-1.40

cting

-1.39

 khusus

-1.39

POSITIVE LOGITS

“

1.41

全新

1.41

hline

1.39

 .”

1.38

();

1.34

 even

1.32

ст

1.30

 հղումներ

1.30

!”

1.30

""

1.29

Activations Density 0.032%