INDEX

Explanations

people advocating or being

The neuron detects generic references to groups or collectives—words like “those,” “people,” “elements,” or “who” that introduce a group or category.

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

 również

-1.43

 канди

-1.40

photographe

-1.38

 terbang

-1.38

Atelier

-1.34

 dużo

-1.28

MacBook

-1.27

素人

-1.27

 dilaksanakan

-1.26

更何况

-1.26

POSITIVE LOGITS

who

1.72

are

1.72

complished

1.71

 herbes

1.48

อีก

1.44

chromatic

1.35

giv

1.30

ditionally

1.27

structing

1.26

自我

1.21

Activations Density 0.093%