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

    Neuronpedia

    INDEX
    Explanations

    neutral

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    occ
    -0.09
     illustrated
    -0.08
     книг
    -0.08
     يوس
    -0.08
    奄
    -0.07
     רוב
    -0.07
     QApplication
    -0.07
    _bus
    -0.07
     ski
    -0.07
    vg
    -0.07
    POSITIVE LOGITS
     neutrality
    0.08
     favourable
    0.07
     neutral
    0.07
    Gradient
    0.07
    𫞩
    0.07
    谈到
    0.07
    Channel
    0.07
     משמעותי
    0.06
    MONTH
    0.06
    شروط
    0.06
    Act Density 0.009%

    No Known Activations