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

    Neuronpedia

    INDEX
    Explanations

    knowing or not knowing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ج
    0.80
    0.75
    <0x0D>
    0.73
    ك
    0.73
    ش
    0.71
    кту
    0.68
     \
    0.59
    y
    0.59
    م
    0.59
     Vielleicht
    0.59
    POSITIVE LOGITS
    의
    0.81
    _
    0.72
    の
    0.71
    ır
    0.71
     که
    0.69
    ه‌های
    0.68
    すぎる
    0.67
     berjudul
    0.66
    sey
    0.65
     dobrze
    0.65
    Act Density 0.060%

    No Known Activations