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

    Neuronpedia

    INDEX
    Explanations

    importance and correctness

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     mampu
    0.46
     capables
    0.43
     말미암
    0.43
    டைந்து
    0.40
     შეუ
    0.40
     läh
    0.39
    ⤥
    0.39
     steve
    0.39
    zetac
    0.39
    极大
    0.39
    POSITIVE LOGITS
     måste
    0.80
     correctly
    0.77
     beachten
    0.73
    必须
    0.72
    要注意
    0.72
     careful
    0.71
    一定要
    0.71
     jangan
    0.70
    注意
    0.69
     반드시
    0.69
    Act Density 0.359%

    No Known Activations