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

    Neuronpedia

    INDEX
    Explanations

    half hours

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ↵
    0.77
    ا
    0.75
     odor
    0.75
     organise
    0.73
     aroma
    0.71
    L
    0.70
     of
    0.70
    移民
    0.69
     al
    0.68
     organ
    0.68
    POSITIVE LOGITS
     Ты
    0.93
     кры
    0.91
     когда
    0.84
     применение
    0.84
     такие
    0.83
    FindingsResponse
    0.82
     ты
    0.81
     ничего
    0.81
     Бы
    0.80
     опыт
    0.80
    Act Density 0.001%

    No Known Activations