INDEX
    Explanations

    words related to film details, including aspects of storytelling and character traits

    New Auto-Interp
    Negative Logits
    ire
    -0.16
     several
    -0.16
     entire
    -0.16
     certain
    -0.15
    isches
    -0.15
     Pis
    -0.14
     same
    -0.14
    era
    -0.14
    _different
    -0.14
     little
    -0.13
    POSITIVE LOGITS
    -ÑĤо
    0.15
    InThe
    0.15
     же
    0.15
     ترÛĮÙĨ
    0.15
    ENCHMARK
    0.15
    heimer
    0.14
    isyon
    0.14
    mente
    0.14
    azon
    0.13
     поба
    0.13
    Act Density 0.102%

    No Known Activations