INDEX

Explanations

preposition followed by article or noun

This neuron strongly activates on common English prepositions (like “from,” “into,” “on,” and “in”) that introduce relationships or context.

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

 that

-2.47

of

-1.95

-1.69

 using

-1.67

 made

-1.54

 make

-1.52

 should

-1.49

 они

-1.48

 allowing

-1.46

 тот

-1.45

POSITIVE LOGITS

骉

1.86

quelles

1.70

⼼

1.69

zembro

1.57

 différents

1.57

𓁹

1.56

 Reduces

1.55

ۂ

1.55

zelfde

1.53

我们就

1.53

Activations Density 0.208%