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

    Neuronpedia

    INDEX
    Explanations

    unlikely positive or complex statements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    style
    0.62
     a
    0.55
    dan
    0.54
    ser
    0.54
    son
    0.54
    lock
    0.54
    dar
    0.54
    dynamic
    0.54
    value
    0.52
    merk
    0.52
    POSITIVE LOGITS
     différences
    0.47
     Hospitals
    0.46
     hospitales
    0.45
     banca
    0.45
    療法
    0.44
     Batik
    0.44
    績
    0.44
     रवाना
    0.42
    分开
    0.42
     Marne
    0.42
    Act Density 0.001%

    No Known Activations