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

    Neuronpedia

    INDEX
    Explanations

    representing how something is misunderstood

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     skorzyst
    0.77
     utiliza
    0.72
    uses
    0.71
     require
    0.69
     korzyst
    0.67
     använder
    0.66
     uses
    0.66
     usar
    0.65
     keuntungan
    0.65
     utilizzare
    0.65
    POSITIVE LOGITS
     representing
    2.97
    representing
    2.77
     represent
    2.70
     reflecting
    2.65
     represents
    2.61
    代表
    2.52
     reflects
    2.50
     Represents
    2.49
     Represent
    2.47
     reflect
    2.45
    Act Density 0.844%

    No Known Activations