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

    Neuronpedia

    INDEX
    Explanations

    floating-point numbers and precision

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     MISSION
    0.40
    сько
    0.39
    沸
    0.39
     ARTICLE
    0.38
    husband
    0.38
     टाकी
    0.38
    ოლოდ
    0.38
    bsen
    0.38
    ொள்ள
    0.37
    omt
    0.37
    POSITIVE LOGITS
    Floating
    0.69
     floating
    0.68
     Floating
    0.68
     Flo
    0.63
    parseFloat
    0.59
     FLO
    0.58
     flott
    0.57
     float
    0.54
    Flo
    0.54
    浮
    0.54
    Act Density 0.015%

    No Known Activations