INDEX
    Explanations

    references to playful or whimsical imagery

    New Auto-Interp
    Negative Logits
    pery
    -0.17
    imar
    -0.16
    alse
    -0.15
     Sı
    -0.14
    GuidId
    -0.14
    idth
    -0.14
     CharSequence
    -0.14
    GIN
    -0.14
     nÄĥ
    -0.13
    à¸Ńà¸Ķ
    -0.13
    POSITIVE LOGITS
     hearts
    0.22
     abstract
    0.19
    .Abstract
    0.18
    abstract
    0.18
     heart
    0.18
     flowers
    0.18
     leaves
    0.17
     flames
    0.17
     bear
    0.17
     lady
    0.16
    Act Density 0.213%

    No Known Activations