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

    Neuronpedia

    INDEX
    Explanations

    I tell you learn

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     only
    0.88
     somente
    0.81
    only
    0.79
    可以
    0.75
     requires
    0.73
     necesarios
    0.71
     Only
    0.70
     apenas
    0.68
     лишь
    0.68
     necesarias
    0.68
    POSITIVE LOGITS
    Kết
    0.58
    కుంది
    0.57
    Номер
    0.57
     בא
    0.56
    Career
    0.55
     پ
    0.54
     giornata
    0.54
    ském
    0.54
     кең
    0.53
    workout
    0.53
    Act Density 0.051%

    No Known Activations