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

    Neuronpedia

    INDEX
    Explanations

    primarily

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     pierws
    0.78
    一段时间
    0.77
     particolarmente
    0.76
     primera
    0.76
     něk
    0.73
     caso
    0.73
     potrebno
    0.73
    が多く
    0.73
    ");
    0.72
    boxylic
    0.71
    POSITIVE LOGITS
    ileges
    0.75
    s
    0.70
    ਆਂ
    0.69
    та
    0.68
     havoc
    0.68
    с
    0.68
     composure
    0.67
    心思
    0.66
    hetics
    0.65
    ς
    0.65
    Act Density 0.002%

    No Known Activations