INDEX
    Explanations

    voice is/are, inputs is/are, characters is/are

    New Auto-Interp
    Negative Logits
    0.76
    ázky
    0.72
     жі
    0.72
    解消
    0.70
     жүктөп
    0.64
     ജോലി
    0.64
     الصغيرة
    0.63
    izmus
    0.63
    зни
    0.63
    বন্দ
    0.63
    POSITIVE LOGITS
     commensurate
    0.75
     deviated
    0.74
     hydrated
    0.72
     shuffled
    0.70
     transcendent
    0.69
     quantized
    0.69
     helt
    0.68
     differentiated
    0.68
    formal
    0.67
     switched
    0.67
    Act Density 0.652%

    No Known Activations