INDEX
    Explanations

    This neuron activates on the indefinite article “an.”

    New Auto-Interp
    Negative Logits
    ñana
    -0.06
     Cheese
    -0.06
     yoktur
    -0.06
    -0.06
     fries
    -0.06
     truyền
    -0.06
     convex
    -0.06
    dbName
    -0.06
     lợi
    -0.06
    asename
    -0.06
    POSITIVE LOGITS
     ст
    0.07
     grids
    0.06
    _MARGIN
    0.06
    ­ing
    0.06
     abruptly
    0.06
    anson
    0.06
     fascinated
    0.06
     та
    0.06
     баг
    0.06
     مساحت
    0.06
    Act Density 0.032%

    No Known Activations