Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    object orientation and size

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     среднем
    0.49
     obce
    0.47
     literalmente
    0.46
     जीवित
    0.45
     Driven
    0.44
     Reason
    0.44
     Translated
    0.44
     postpaid
    0.44
     առաջ
    0.43
    ଞ
    0.43
    POSITIVE LOGITS
     spacing
    1.15
     shape
    1.12
     placement
    1.08
     layout
    1.06
     angles
    0.99
     orientation
    0.98
     angle
    0.98
     configuration
    0.96
     sizing
    0.96
     positioning
    0.96
    Act Density 0.416%

    No Known Activations