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

    Neuronpedia

    INDEX
    Explanations

    numerical scales and ratings

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    淳
    1.23
    ギュ
    1.03
     사회
    0.95
     FILES
    0.95
    ENGINE
    0.93
    純
    0.93
    永遠
    0.92
     DEATH
    0.92
    诵
    0.92
    レント
    0.92
    POSITIVE LOGITS
    というか
    1.23
    ovski
    1.18
     gezien
    1.17
    рила
    1.09
     respe
    1.07
     самостоя
    1.00
    gable
    0.98
    packet
    0.98
    рил
    0.98
     algo
    0.98
    Act Density 0.030%

    No Known Activations