INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    ieur
    -0.08
    936
    -0.08
     शिक्षक
    -0.08
    ivation
    -0.08
     rằng
    -0.08
     i've
    -0.07
     disting
    -0.07
     निर्ध
    -0.07
    339
    -0.07
     परिस
    -0.07
    POSITIVE LOGITS
    half
    0.12
     half
    0.11
    Half
    0.11
     Half
    0.11
     metade
    0.10
     metà
    0.10
     نصف
    0.10
     halve
    0.10
    _half
    0.10
     moitié
    0.10
    Act Density 0.034%

    No Known Activations