INDEX
    Explanations

    width x height dimensions

    New Auto-Interp
    Negative Logits
     induction
    0.37
    lasso
    0.36
     struck
    0.35
     Iglesia
    0.35
    Watson
    0.35
     upper
    0.35
    ‌اند
    0.34
    bud
    0.34
     spurt
    0.34
     leaf
    0.33
    POSITIVE LOGITS
    Resolution
    0.44
    Pixels
    0.44
     його
    0.42
     pixels
    0.42
     திரை
    0.42
     его
    0.42
     Resolución
    0.41
     Resolution
    0.41
    分辨率
    0.40
     перемо
    0.40
    Act Density 0.006%

    No Known Activations