INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     facets
    -0.08
     samen
    -0.08
    hearted
    -0.07
     تلك
    -0.07
     funnels
    -0.07
     perfectly
    -0.07
     செய்வ
    -0.07
     funnel
    -0.07
     salah
    -0.07
    ooq
    -0.07
    POSITIVE LOGITS
     Jules
    0.10
     Jump
    0.10
     jumps
    0.09
     Kick
    0.09
    คือ
    0.08
     approx
    0.08
     jumped
    0.08
    rical
    0.08
     없음
    0.08
    0.08
    Act Density 0.085%

    No Known Activations