INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     homework
    -0.08
    培养
    -0.07
     subtly
    -0.07
    .choice
    -0.07
    .quit
    -0.07
    .Text
    -0.07
    586
    -0.07
    auto
    -0.07
     beautifully
    -0.07
     brilliantly
    -0.07
    POSITIVE LOGITS
     kilometer
    0.10
     displaced
    0.09
     hollow
    0.09
     лаз
    0.09
     bort
    0.09
     cylindrical
    0.09
     voxel
    0.09
     centimeter
    0.09
     spherical
    0.09
     bale
    0.09
    Act Density 0.010%

    No Known Activations