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

    Neuronpedia

    INDEX
    Explanations

    aesthetic explanations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    не
    2.28
    ある
    1.77
    أ
    1.73
    Ｄ
    1.73
    нг
    1.70
    బాద్
    1.59
    де
    1.59
    ำ
    1.58
    يد
    1.52
    бо
    1.52
    POSITIVE LOGITS
    etheless
    1.98
    theless
    1.83
    ോട്ട്
    1.82
    ate
    1.77
    så
    1.75
    selling
    1.73
    데요
    1.71
    s
    1.69
    ことにより
    1.66
    meter
    1.65
    Act Density 0.002%

    No Known Activations