INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    кус
    -0.08
    anne
    -0.08
    nummer
    -0.07
    ков
    -0.07
    éis
    -0.07
    raising
    -0.07
    raise
    -0.07
     rinse
    -0.07
     нанес
    -0.07
     Boards
    -0.07
    POSITIVE LOGITS
     canopy
    0.10
     foliage
    0.09
    Dense
    0.09
     dense
    0.09
     jungle
    0.08
    0.08
    0.08
     Dense
    0.08
    0.08
    0.08
    Act Density 0.006%

    No Known Activations