INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     হৃদ
    -0.08
    -0.08
     camin
    -0.08
     انسان
    -0.08
    ahlt
    -0.08
    ғыс
    -0.08
    -0.08
    ده
    -0.07
     caminhada
    -0.07
     jubile
    -0.07
    POSITIVE LOGITS
     everywhere
    0.08
     inconsist
    0.08
    शी
    0.08
     wherever
    0.08
     scr
    0.08
    -ath
    0.08
     incons
    0.07
     reaffirm
    0.07
     updating
    0.07
     essayer
    0.07
    Act Density 0.005%

    No Known Activations