INDEX

Explanations

investment and return

The neuron activates on numeric financial return metrics—especially ROI figures and percentage-based investment returns.

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

biendo

-1.68

lenül

-1.64

р

-1.60

饽

-1.47

 deswegen

-1.43

 llegan

-1.41

zzie

-1.41

"};

-1.41

%");

-1.38

to

-1.37

POSITIVE LOGITS

—

1.69

zkę

1.50

うえ

1.47

of

1.45

長の

1.43

1.40

 μπο

1.36

ственных

1.36

逦

1.33

 estan

1.32

Activations Density 0.003%