INDEX

Explanations

hourly rates or frequency

This neuron detects references to payment terms—especially “hourly” (and related wage/rate) or salaried compensation.

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

But

-3.00

 According

-2.70

ܜ

-2.42

</h5>

-2.33

 και

-2.27

 fondamentale

-2.22

$};

-2.22



-2.16

AND

-2.14

 Algunos

-2.13

POSITIVE LOGITS

of

3.34

2.94

𞤢

2.75

</strong>

2.69

 этих

2.66

,--

2.52

2.47

2.42

dzić

2.36

𐐀

2.36

Activations Density 0.002%