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

    Neuronpedia

    INDEX
    Explanations

    transform(-50%, -50%)

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ра
    0.91
    ي
    0.89
    rául
    0.82
    ি
    0.76
    ار
    0.73
    йся
    0.70
    тивы
    0.69
    ᅲ
    0.69
    État
    0.68
    ствую
    0.67
    POSITIVE LOGITS
     nombr
    0.84
     tokamak
    0.77
     ⠀
    0.75
     Đ
    0.74
    тинен
    0.72
     ⚫
    0.71
     liderazgo
    0.71
     ⋅
    0.71
     derive
    0.70
     ‫
    0.70
    Act Density 0.001%

    No Known Activations