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

    Neuronpedia

    INDEX
    Explanations

    terrible or poor evaluations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    উট
    0.92
    ${
    0.92
    tras
    0.90
    Successfully
    0.89
    cate
    0.89
     sheer
    0.87
    <unused2103>
    0.86
    크
    0.85
     junctions
    0.84
    trp
    0.84
    POSITIVE LOGITS
    س
    1.28
    гона
    1.09
    وب
    1.01
     geeignet
    0.99
     खासा
    0.96
     möglich
    0.93
    anı
    0.93
     ইসলামাবাদ
    0.93
     durumda
    0.92
    д
    0.91
    Act Density 0.030%

    No Known Activations