INDEX

Explanations

is boolean properties

The neuron activates on identifiers or attribute names beginning with “is” (e.g. boolean flags like IsSuccess, is_active, IsReadOnly).

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

などを

-1.63

 with

-1.57

 czter

-1.52

 kwiet

-1.48

-1.45

 portavoz

-1.44

 réduite

-1.41

 aportes

-1.38

 nový

-1.38

 hermosos

-1.38

POSITIVE LOGITS

</h1>

1.59

]}>

1.51

1.45

Inhalts

1.38

</b>

1.36

it

1.35

1.34

żs

1.34

 katk

1.29

 poffible

1.29

Activations Density 0.021%