INDEX
    Explanations

    sentences about familial relationships and expressions of personal identity

    New Auto-Interp
    Negative Logits
     spirit
    -0.15
    imb
    -0.14
    ADED
    -0.14
    erie
    -0.14
    Priv
    -0.14
    alim
    -0.14
    iale
    -0.14
    404
    -0.14
    imas
    -0.14
    feld
    -0.13
    POSITIVE LOGITS
     Mud
    0.16
    xffffffff
    0.16
    uell
    0.16
     Nuclear
    0.15
    견
    0.15
    ож
    0.14
    DY
    0.14
    anko
    0.14
     Spl
    0.14
    egen
    0.14
    Act Density 0.080%

    No Known Activations