INDEX
    Explanations

    evaluative phrases that discuss film quality and structure

    New Auto-Interp
    Negative Logits
    hea
    -0.07
    ît
    -0.07
     Kend
    -0.06
    urtles
    -0.06
    now
    -0.06
    bable
    -0.06
    ysz
    -0.06
    èĬĿ
    -0.06
    .opensource
    -0.06
    amework
    -0.06
    POSITIVE LOGITS
     Overall
    0.08
     overall
    0.07
    Overall
    0.07
    orny
    0.06
    CurrentValue
    0.06
    overall
    0.06
    LBL
    0.06
    aben
    0.06
    jav
    0.06
     speaking
    0.06
    Act Density 0.021%

    No Known Activations