INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     essence
    0.72
     e
    0.72
     (
    0.71
     utility
    0.71
     workpiece
    0.69
     utilitarian
    0.68
     you
    0.68
     Frankenstein
    0.68
     products
    0.67
     Calculus
    0.66
    POSITIVE LOGITS
    SizeMode
    0.99
    ającą
    0.97
    允许
    0.97
    Mode
    0.95
    Type
    0.93
    也不
    0.93
    ertype
    0.93
     během
    0.93
    ónicos
    0.93
    Allow
    0.92
    Act Density 0.439%

    No Known Activations