INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     Rate
    -0.08
     uiteen
    -0.08
    fst
    -0.08
    hic
    -0.08
     אב
    -0.08
     پھ
    -0.08
    τήσεις
    -0.08
     фрукт
    -0.07
    rolley
    -0.07
     proximity
    -0.07
    POSITIVE LOGITS
     congr
    0.29
     equal
    0.27
    equal
    0.22
    _equal
    0.21
    Equal
    0.21
     Cong
    0.21
    Cong
    0.20
    _EQUAL
    0.20
     рав
    0.20
     Equal
    0.20
    Act Density 0.072%

    No Known Activations