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

    Neuronpedia

    INDEX
    Explanations

    committee mission and members

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     that
    0.88
    män
    0.88
    t
    0.78
    তে
    0.76
    ка
    0.75
    че
    0.73
    that
    0.72
     can
    0.71
    commission
    0.71
     be
    0.71
    POSITIVE LOGITS
    ᱓
    0.81
     maneras
    0.80
    あらゆる
    0.79
    ޘ
    0.76
     XGB
    0.75
    燹
    0.74
    $}
    0.73
    خراج
    0.72
    '}
    0.71
    el
    0.70
    Act Density 0.001%

    No Known Activations