INDEX
    Explanations

    references to time and duration, particularly in hours

    New Auto-Interp
    Negative Logits
    olly
    -0.17
    ollipop
    -0.17
    ohn
    -0.16
    imes
    -0.16
    anzi
    -0.15
    builder
    -0.15
    illions
    -0.14
    acus
    -0.14
    ingroup
    -0.14
    guard
    -0.14
    POSITIVE LOGITS
     ÑĩаÑģа
    0.16
    â̳
    0.15
    ously
    0.14
    -hour
    0.14
    kees
    0.14
    Äįem
    0.14
    اÙĨÛĮ
    0.14
    lasting
    0.14
    uates
    0.14
    vey
    0.13
    Act Density 0.065%

    No Known Activations