INDEX
    Explanations

    tokens representing years, dates, or other multi-digit numeric sequences (e.g., "2023", "2015").

    New Auto-Interp
    Negative Logits
     when
    0.60
    when
    0.57
    і
    0.55
    ко
    0.54
     during
    0.51
    One
    0.51
     one
    0.50
     always
    0.49
    it
    0.49
    0.49
    POSITIVE LOGITS
    డీపీ
    0.59
     enquête
    0.58
     défis
    0.57
     únic
    0.56
     יותר
    0.55
     ईमित्र
    0.54
     चुनौतियों
    0.54
    可能是
    0.54
     റിപ്പോ
    0.52
     बेटियों
    0.52
    Act Density 0.082%

    No Known Activations