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

    Neuronpedia

    INDEX
    Explanations

    Let's say examples

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     sowie
    0.74
    があれば
    0.70
    甚至是
    0.68
    更能
    0.67
    すれば
    0.66
     حتی
    0.66
    क्राइब
    0.66
    之外
    0.65
    也能
    0.65
    න්ද
    0.65
    POSITIVE LOGITS
     muốn
    0.96
     want
    0.86
     хочет
    0.84
     decided
    0.83
     상황
    0.82
     đang
    0.81
    Suddenly
    0.80
     ingin
    0.79
     хотят
    0.79
     deseja
    0.78
    Act Density 0.399%

    No Known Activations