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

    Neuronpedia

    INDEX
    Explanations

    not always

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     não
    1.03
     ikke
    0.93
     пусть
    0.88
     nicht
    0.86
     либо
    0.86
    não
    0.86
     ไม่
    0.85
     tidak
    0.82
     אין
    0.82
     скорее
    0.81
    POSITIVE LOGITS
     necessarily
    1.07
    orious
    0.94
    necessarily
    0.92
     uncommon
    0.90
     yet
    0.89
     negotiable
    0.89
     advisable
    0.89
    orius
    0.88
     always
    0.88
     applicable
    0.88
    Act Density 0.356%

    No Known Activations