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

    Neuronpedia

    INDEX
    Explanations

    programming definitions with assorted symbols

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    the
    0.78
    whose
    0.74
    certain
    0.72
     overcrowded
    0.70
    typical
    0.69
    trying
    0.69
    namely
    0.68
    tand
    0.67
    existing
    0.65
    easily
    0.65
    POSITIVE LOGITS
    /
    0.68
     প্রশংসা
    0.67
    _
    0.66
    ное
    0.64
    复
    0.64
    iquet
    0.63
    的目的
    0.61
    种
    0.60
    组件
    0.59
    득
    0.58
    Act Density 0.676%

    No Known Activations