INDEX

Explanations

increasing size or density

The neuron fires on words that signal technical or quantitative descriptors—i.e. measurement, quantity, or numeric‐specification terms (e.g. “point,” “higher,” “frequencies,” “size,” “density,” “rates”).

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

 worst

-0.86

そのまま

-0.81

ализация

-0.81

 unprepared

-0.80

 fonction

-0.80

悠

-0.79

pâ

-0.78

正好

-0.76

 gobernador

-0.76

 didn

-0.75

POSITIVE LOGITS

 ormai

1.72

 artık

1.31

 increasingly

1.23

 deviennent

1.18

 anymore

1.16

もはや

1.13

 becomes

1.05

 naprawdę

1.05

 remaining

1.03

 mittlerweile

1.03

Activations Density 0.058%