INDEX

Explanations

possessives, common nouns

The neuron strongly activates on words that signal relationships or attributes—especially prepositions like “with,” “to,” “in,” “as,” “for,” and related relational nouns (e.g. “year,” “degree,” “ability”).

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

the

-1.39

 никогда

-1.31

Our

-1.31

ورة

-1.27

meldet

-1.22

ECTOS

-1.13

ずっと

-1.11

 thấy

-1.10

 nunca

-1.09

有些人

-1.09

POSITIVE LOGITS

its

2.00

his

1.77

 jego

1.59

who

1.57

 Consejos

1.42

 ње

1.38

 его

1.34

 lakini

1.34

 שלו

1.30

belum

1.28

Activations Density 0.340%