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

    Neuronpedia

    INDEX
    Explanations

    despite the + contrasting noun

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    icking
    0.83
    ত্ত্ব
    0.81
    وٹ
    0.78
    denly
    0.75
    ৮
    0.73
    ление
    0.73
    ৫
    0.73
    ilight
    0.71
    鼐
    0.71
    ri
    0.70
    POSITIVE LOGITS
    그러나
    0.91
    ↵
    0.88
     그러나
    0.84
     vậy
    0.80
    ات
    0.77
    但是在
    0.77
    ्स
    0.76
    새
    0.73
    但是
    0.72
     AIDS
    0.71
    Act Density 0.000%

    No Known Activations