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

    Neuronpedia

    INDEX
    Explanations

    Marrakech activities and descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     možda
    1.70
    za
    1.64
    aklar
    1.60
    诰
    1.57
    ak
    1.55
     trái
    1.52
     teilweise
    1.52
    á
    1.52
    or
    1.48
    ata
    1.47
    POSITIVE LOGITS
    на
    1.95
    ان
    1.94
    有助于
    1.88
    ן
    1.76
    િંગ
    1.73
    ब्दिक
    1.70
    이션
    1.70
    有利于
    1.69
     curtains
    1.66
     ampli
    1.64
    Act Density 0.000%

    No Known Activations