INDEX
    Explanations

    mentions of specific holidays, such as Thanksgiving and Valentine's Day

    references to Thanksgiving and related holidays

    New Auto-Interp
    Negative Logits
    ially
    -0.88
     constitu
    -0.87
    ioch
    -0.84
    ials
    -0.76
    upon
    -0.74
    etting
    -0.71
    umbn
    -0.69
     defin
    -0.69
    erd
    -0.69
    iological
    -0.68
    POSITIVE LOGITS
     eve
    0.97
     Day
    0.95
     Eve
    0.92
     festivities
    0.86
     Thanksgiving
    0.84
     Month
    0.83
    nesday
    0.83
     holidays
    0.82
     dinner
    0.82
     Surprise
    0.81
    Act Density 0.012%

    No Known Activations