INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     the
    0.91
     a
    0.80
     long
    0.79
     so
    0.77
     any
    0.76
     just
    0.75
    0.74
     not
    0.74
     some
    0.74
     an
    0.73
    POSITIVE LOGITS
    ovič
    0.77
    ović
    0.75
    kowski
    0.75
    ynski
    0.74
    ković
    0.74
    ovski
    0.73
    ißler
    0.72
    ström
    0.72
    nić
    0.71
    opoulos
    0.70
    Act Density 0.018%

    No Known Activations