INDEX
    Explanations

    proper nouns related to names and titles

    New Auto-Interp
    Negative Logits
    emmel
    -0.15
    ocab
    -0.15
    ocado
    -0.14
     cont
    -0.14
    éļ
    -0.14
    ãģ«ãģ¦
    -0.13
    mÄĽ
    -0.13
    Unnamed
    -0.13
    åı£
    -0.13
    führ
    -0.13
    POSITIVE LOGITS
     wasn
    0.16
     isn
    0.16
    amin
    0.15
    asma
    0.15
     lamin
    0.15
    akin
    0.14
     Dit
    0.14
     aren
    0.14
    ool
    0.14
     uns
    0.14
    Act Density 0.132%

    No Known Activations