INDEX
    Explanations

    phrases conveying degrees of comparison or emphasis, particularly involving the word "far."

    New Auto-Interp
    Negative Logits
    AutoScaleMode
    -0.67
     onely
    -0.62
     gustó
    -0.61
    gabe
    -0.60
    printList
    -0.59
    engager
    -0.59
     Schaefer
    -0.59
     Rosal
    -0.59
    doit
    -0.57
    [`
    -0.55
    POSITIVE LOGITS
    far
    1.06
     far
    0.98
    раздо
    0.94
     FAR
    0.90
    FAR
    0.89
    away
    0.88
    Away
    0.83
    urably
    0.79
    AWAY
    0.79
     away
    0.78
    Act Density 0.107%

    No Known Activations