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

    Neuronpedia

    INDEX
    Explanations

    abstract concepts described

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Because
    1.03
    porque
    1.00
    because
    0.98
     weil
    0.94
    Porque
    0.93
     Porque
    0.92
     Потому
    0.87
     perché
    0.87
     Because
    0.86
     Çünkü
    0.86
    POSITIVE LOGITS
     involving
    1.71
     which
    1.53
     whereby
    1.47
     resulting
    1.44
     culminating
    1.39
     caused
    1.39
     amounting
    1.29
     resembling
    1.28
    which
    1.28
     والتي
    1.27
    Act Density 0.088%

    No Known Activations