INDEX
    Explanations

    This neuron fires on words describing the referee’s or official’s signal to start (or resume) a game or contest.

    New Auto-Interp
    Negative Logits
    _();↵
    -0.07
    ']}↵
    -0.06
    phyl
    -0.06
     maiden
    -0.06
     Jamal
    -0.06
    unately
    -0.06
    Mount
    -0.06
     cultivated
    -0.06
     znám
    -0.06
     Katz
    -0.06
    POSITIVE LOGITS
     zij
    0.08
     مي
    0.06
    лі
    0.06
    INTEGER
    0.06
    oub
    0.06
     ud
    0.06
    patibility
    0.06
    0.06
    0.06
    rig
    0.06
    Act Density 0.016%

    No Known Activations