INDEX
    Explanations

    mentions of ice cream and related dessert items

    New Auto-Interp
    Negative Logits
     ÅŁar
    -0.17
     bread
    -0.16
     frying
    -0.16
     underwater
    -0.15
    bread
    -0.15
    oise
    -0.15
     bottle
    -0.15
     Marketable
    -0.15
    å¼ķãģį
    -0.15
     cooking
    -0.14
    POSITIVE LOGITS
     ice
    0.57
     Ice
    0.48
    ice
    0.45
    Ice
    0.42
    åĨ°
    0.38
     ICE
    0.38
     scoop
    0.37
     frozen
    0.35
     icy
    0.34
     sco
    0.33
    Act Density 0.034%

    No Known Activations