INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     bot
    -0.07
     contenant
    -0.07
     बेटे
    -0.07
     métal
    -0.07
    غير
    -0.07
     envia
    -0.07
     flam
    -0.07
     procedure
    -0.07
     dirinya
    -0.07
     shear
    -0.07
    POSITIVE LOGITS
    Base
    0.08
     básico
    0.08
     Rc
    0.08
    Rc
    0.08
    entesque
    0.08
    基本
    0.08
    Throughout
    0.08
     Throughout
    0.08
    -base
    0.08
    .base
    0.07
    Act Density 0.004%

    No Known Activations