INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     выбира
    0.54
     raped
    0.54
     длину
    0.54
     iniciando
    0.52
     размер
    0.51
     ejected
    0.50
     died
    0.50
     longueur
    0.50
     использовали
    0.50
     gave
    0.48
    POSITIVE LOGITS
     কেননা
    0.66
    Meta
    0.65
        
    0.64
     Because
    0.62
     Notably
    0.61
    Because
    0.61
     Though
    0.61
    Though
    0.60
    Anyway
    0.60
     เพราะ
    0.60
    Act Density 0.052%

    No Known Activations