INDEX
    Explanations

    references to geographical locations and physical features

    New Auto-Interp
    Negative Logits
    izr
    -0.18
    izona
    -0.16
    à¥Ģध
    -0.15
    ÑĢив
    -0.14
    acier
    -0.14
    andest
    -0.14
    essian
    -0.14
     sns
    -0.14
     Shore
    -0.14
    ousel
    -0.14
    POSITIVE LOGITS
     island
    0.25
     Island
    0.22
     islands
    0.21
     Islands
    0.20
     unin
    0.19
     оÑģÑĤÑĢов
    0.16
     ostrov
    0.16
    ÑģÑĤÑĢов
    0.15
    pong
    0.15
    å³¶
    0.15
    Act Density 0.094%

    No Known Activations