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

    Neuronpedia

    INDEX
    Explanations

    organize this

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     možné
    1.86
    ↵
    1.85
    がつ
    1.80
     중
    1.73
    合い
    1.73
     וכ
    1.73
    ort
    1.71
    নি
    1.71
    ১৯
    1.66
     מח
    1.66
    POSITIVE LOGITS
    ى
    2.94
    ת
    2.38
    ع
    2.17
    liness
    2.16
    vres
    2.00
    يين
    1.97
    vents
    1.95
    callbacks
    1.93
    de
    1.92
    puff
    1.90
    Act Density 0.138%

    No Known Activations