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

    Neuronpedia

    INDEX
    Explanations

    used to, received by

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     --
    0.82
     ½
    0.72
     –
    0.71
     roared
    0.70
     world
    0.69
     ‘
    0.68
     heath
    0.67
     conti
    0.67
     ·
    0.65
     x
    0.65
    POSITIVE LOGITS
     asimismo
    0.97
     XNUMX
    0.95
     möglicherweise
    0.91
    咱们
    0.78
    скольку
    0.77
     yalnızca
    0.75
     Поскольку
    0.74
     поскольку
    0.73
    либо
    0.72
     এটির
    0.71
    Act Density 0.049%

    No Known Activations