INDEX
    Explanations

    movie genres

    The neuron activates on the genre‐style viewer type labels (e.g. “Comedy,” “Action,” “Fantasy,” etc.) in the list of viewer types.

    New Auto-Interp
    Negative Logits
    lardı
    -0.07
     harmonic
    -0.07
    Dock
    -0.07
     APK
    -0.07
    Flutter
    -0.06
    BST
    -0.06
     деревян
    -0.06
     눈을
    -0.06
    ırlar
    -0.06
    aks
    -0.06
    POSITIVE LOGITS
     eh
    0.07
    ايش
    0.07
    важа
    0.06
     consolid
    0.06
     demol
    0.06
     CONFIG
    0.06
    quarter
    0.06
    _SITE
    0.06
     tonnes
    0.06
    0.06
    Act Density 0.008%

    No Known Activations