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

    Neuronpedia

    INDEX
    Explanations

    millions and billions scale

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     sıcak
    0.80
    acketing
    0.77
     будете
    0.76
     angles
    0.75
    reetype
    0.75
     angled
    0.74
     乀
    0.74
     ángulos
    0.73
     கரைச
    0.73
     ہمیشہ
    0.72
    POSITIVE LOGITS
     millions
    2.15
     thousands
    2.08
     billions
    1.95
     Millions
    1.89
     hundreds
    1.87
     Thousands
    1.80
    Millions
    1.78
    Thousands
    1.75
    thousands
    1.74
     million
    1.73
    Act Density 0.443%

    No Known Activations