INDEX
    Explanations

    phrases emphasizing affiliation or belonging in a community or group context

    New Auto-Interp
    Negative Logits
    身ä¸Ĭ
    -0.17
    zn
    -0.16
    116
    -0.15
    enga
    -0.15
    zac
    -0.15
    ets
    -0.14
    lea
    -0.14
    ics
    -0.13
    endas
    -0.13
     einzel
    -0.13
    POSITIVE LOGITS
     larger
    0.26
     wider
    0.22
     Larger
    0.22
    larg
    0.21
     broader
    0.19
     bigger
    0.19
    -package
    0.19
     ongoing
    0.19
     normal
    0.18
     package
    0.18
    Act Density 0.043%

    No Known Activations