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

    Neuronpedia

    INDEX
    Explanations

    stylesheet links in HTML

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    лец
    -0.81
    mrow
    -0.81
    アニメ
    -0.81
    户
    -0.79
     कोण
    -0.79
     eaux
    -0.76
    laa
    -0.75
    RQ
    -0.74
    Exercise
    -0.72
     Elo
    -0.71
    POSITIVE LOGITS
    œur
    0.87
    stylesheet
    0.85
    oltán
    0.78
    Both
    0.77
    quiler
    0.77
     after
    0.76
    要不是
    0.76
     [&
    0.75
     beak
    0.74
     pemas
    0.74
    Act Density 0.010%

    No Known Activations