INDEX
    Explanations

    mentions of recipes and cooking-related terms

    New Auto-Interp
    Negative Logits
     Rand
    -0.52
     “
    -0.49
    under
    -0.49
     Whittaker
    -0.48
     Hull
    -0.47
    mau
    -0.46
     Савезне
    -0.46
     שוליים
    -0.45
     North
    -0.45
     Williams
    -0.44
    POSITIVE LOGITS
     recipe
    1.90
     Recipe
    1.59
    recipe
    1.50
    Recipe
    1.50
     recipes
    1.49
     RECIPE
    1.49
     Recipes
    1.32
     receta
    1.21
    recipes
    1.16
     recette
    1.15
    Act Density 0.001%

    No Known Activations