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

    Neuronpedia

    INDEX
    Explanations

    digits and special characters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     darn
    0.79
     viejo
    0.76
    老化
    0.71
     signboard
    0.71
     cigarette
    0.71
     venerable
    0.70
     Jamb
    0.69
     Bajo
    0.69
     old
    0.68
     রহিল
    0.68
    POSITIVE LOGITS
     refroid
    0.66
    lis
    0.66
    nt
    0.65
    ita
    0.61
    uelle
    0.61
    綺麗
    0.60
     stoff
    0.60
    bine
    0.60
     florida
    0.59
    ైనా
    0.58
    Act Density 0.189%

    No Known Activations