INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    conexion
    -0.79
     sarili
    -0.79
    ualmente
    -0.72
    imageURL
    -0.72
    コンセプト
    -0.71
    cimentos
    -0.69
     συγ
    -0.68
    -0.68
    λικ
    -0.68
     ancest
    -0.66
    POSITIVE LOGITS
     letter
    1.13
     ana
    1.13
    Ratio
    1.09
    idescreen
    1.05
     pillar
    1.01
     aspect
    0.96
     framing
    0.95
     ratio
    0.93
    letter
    0.93
     Ratios
    0.92
    Act Density 0.026%

    No Known Activations