INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     herself
    0.48
     mom
    0.46
     elevada
    0.45
     wrecked
    0.44
    野菜
    0.43
     alumna
    0.42
     grandma
    0.42
     helps
    0.41
    یکه
    0.41
    0.41
    POSITIVE LOGITS
     Father
    0.68
    Father
    0.68
    father
    0.67
     father
    0.66
     patern
    0.64
     отец
    0.57
    fathers
    0.55
     아버지
    0.55
     paternal
    0.54
     padre
    0.53
    Act Density 0.004%

    No Known Activations