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

    Neuronpedia

    INDEX
    Explanations

    losing things or people

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    би
    0.62
    ки
    0.61
    리
    0.55
    2
    0.55
    но
    0.54
    gegen
    0.53
    ،
    0.52
    يز
    0.52
    ня
    0.50
    wärts
    0.50
    POSITIVE LOGITS
     lost
    0.53
     it
    0.47
     the
    0.46
     Medicaid
    0.45
     DARK
    0.44
     CH
    0.42
     us
    0.42
     n
    0.41
     pea
    0.41
    る
    0.40
    Act Density 0.010%

    No Known Activations