INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     imaginary
    -0.09
    нике
    -0.08
    怎么办
    -0.08
     टू
    -0.08
    ões
    -0.08
     தொ
    -0.08
    îne
    -0.07
    áter
    -0.07
    urses
    -0.07
     inexp
    -0.07
    POSITIVE LOGITS
     hoped
    0.12
     aims
    0.12
    0.11
     aiming
    0.11
     intends
    0.10
     hopes
    0.10
     sollen
    0.09
     hope
    0.09
     ambitious
    0.09
    ,希望
    0.09
    Act Density 0.153%

    No Known Activations