INDEX
    Explanations

    mentions of different kinds of nuts, with a strong preference for almonds

    mentions of specific places and food items, particularly almond-related products

    New Auto-Interp
    Negative Logits
    enegger
    -0.87
    ledged
    -0.86
    achu
    -0.79
    nown
    -0.76
    kie
    -0.76
    liness
    -0.74
    ledge
    -0.72
    pload
    -0.71
    ners
    -0.71
    ifying
    -0.70
    POSITIVE LOGITS
    onso
    0.90
    ibaba
    0.78
    manac
    0.77
    aline
    0.72
    gebra
    0.70
    asca
    0.69
    chemy
    0.69
    stadt
    0.69
    udic
    0.69
    enic
    0.69
    Act Density 0.052%

    No Known Activations