INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     resolução
    -1.06
    1
    -1.02
     portavoz
    -0.97
     marvelous
    -0.96
     sloppy
    -0.94
     préstamo
    -0.93
     culoare
    -0.92
     gebra
    -0.92
     horrid
    -0.92
    lussen
    -0.91
    POSITIVE LOGITS
     conversa
    1.45
     starters
    1.39
    会話
    1.34
     Conversation
    1.33
     conversation
    1.32
     conversación
    1.26
    conversation
    1.17
     conversational
    1.16
     starter
    1.16
     about
    1.16
    Act Density 0.008%

    No Known Activations