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

    Neuronpedia

    INDEX
    Explanations

    percentages around 50-70%

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ("%
    0.67
     Percent
    0.62
     Officers
    0.59
    Commission
    0.59
    oszt
    0.59
    GOLD
    0.58
    icos
    0.57
     কর্মকর্তাদের
    0.57
    циа
    0.57
    సూరు
    0.57
    POSITIVE LOGITS
    ۔
    0.91
     permut
    0.64
     monochromatic
    0.62
     previs
    0.61
    は
    0.61
     }}
    0.61
     }}$
    0.60
     it
    0.59
     permit
    0.59
     medio
    0.59
    Act Density 0.012%

    No Known Activations