INDEX

Explanations

physical properties of matter

This neuron responds to uncommon, domain-specific nouns—especially technical or specialized terms.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

0.88

0.86

it

0.76

 renovations

0.76

dvs

0.75

á

0.72

an

0.72

 officiating

0.70

the

0.69

POSITIVE LOGITS

 Алла

0.90

 hasonló

0.86

 dangere

0.85

🌖

0.85

 Novgorod

0.83

 Европа

0.80

 Eurasia

0.80

 siguiendo

0.79

Frankreich

0.78

 위험

0.78

Activations Density 0.001%