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

    Neuronpedia

    INDEX
    Explanations

    zuò, Contenders, Grievances

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     can
    0.72
    ী
    0.65
    ik
    0.64
    ५
    0.63
    ↵
    0.61
    त
    0.61
    ை
    0.61
     marché
    0.60
    ing
    0.59
    ير
    0.59
    POSITIVE LOGITS
    0.80
    0
    0.69
    1
    0.63
    srt
    0.55
    lere
    0.54
    stagram
    0.51
     이상의
    0.50
     Abh
    0.50
    quela
    0.48
    ००
    0.47
    Act Density 1.243%

    No Known Activations