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

    Neuronpedia

    INDEX
    Explanations

    punch and holes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    S
    0.97
    ai
    0.95
    า
    0.93
    ک
    0.92
    ল
    0.90
    ও
    0.88
    मी
    0.86
    س
    0.86
    V
    0.81
    ס
    0.79
    POSITIVE LOGITS
    大
    0.86
     punching
    0.68
     Punch
    0.63
     punched
    0.63
     τραγ
    0.63
     отверсти
    0.62
     мали
    0.61
     holes
    0.61
     perfor
    0.59
    rowData
    0.58
    Act Density 0.005%

    No Known Activations