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

    Neuronpedia

    INDEX
    Explanations

    asking user for specific input

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Is
    0.71
     것은
    0.70
    ായത്
    0.68
     Is
    0.68
    KindOf
    0.66
     Doing
    0.65
    zzel
    0.62
     얼마나
    0.62
    isIn
    0.62
     것처럼
    0.61
    POSITIVE LOGITS
     caters
    0.94
     leans
    0.86
     simplifies
    0.84
     focuses
    0.83
     applies
    0.78
     helps
    0.77
    楼主
    0.76
     saves
    0.75
     accommodates
    0.73
     terrestre
    0.73
    Act Density 0.030%

    No Known Activations