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

    Neuronpedia

    INDEX
    Explanations

    object equality and null checks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ודה
    0.44
    嚕
    0.41
     Morphology
    0.38
     carol
    0.38
     thao
    0.38
    ഗോ
    0.38
     Kid
    0.38
    panelMenuList
    0.38
    كتر
    0.37
    ਜ
    0.37
    POSITIVE LOGITS
     equals
    0.77
     equality
    0.77
     Equal
    0.77
     Equals
    0.76
    Equals
    0.74
    Equality
    0.73
     equal
    0.73
     Equality
    0.73
    Equal
    0.70
    equals
    0.69
    Act Density 0.003%

    No Known Activations