INDEX
    Explanations

    references to various patterns and designs

    New Auto-Interp
    Negative Logits
    /apt
    -0.15
    ÌĨ
    -0.14
    remen
    -0.14
    unden
    -0.14
    listener
    -0.13
     Brooks
    -0.13
    ặt
    -0.13
    ذر
    -0.13
    کتر
    -0.13
     Ferd
    -0.13
    POSITIVE LOGITS
     pattern
    0.23
    -pattern
    0.23
     patterns
    0.21
     Pattern
    0.21
     Patterns
    0.20
    (pattern
    0.18
    patterns
    0.17
    .pattern
    0.17
    _pattern
    0.17
    pattern
    0.17
    Act Density 0.114%

    No Known Activations