INDEX
    Explanations

    the word "largely" and sometimes "mostly".

    New Auto-Interp
    Negative Logits
     poitrine
    -0.70
     cheminée
    -0.63
     tarvit
    -0.62
    pure
    -0.60
    wój
    -0.59
     assolu
    -0.59
     duros
    -0.58
    ly
    -0.57
     Pure
    -0.57
     terminée
    -0.57
    POSITIVE LOGITS
     виправивши
    0.85
    脚注の使い方
    0.82
    ichon
    0.82
     Larg
    0.81
     Mous
    0.81
    */,
    0.81
    ']}
    0.81
    /******/
    0.80
    CopyWith
    0.80
    ::::::::::::::::
    0.79
    Act Density 0.006%

    No Known Activations