INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     DSG
    -0.09
    riez
    -0.08
    őr
    -0.08
    flux
    -0.08
     furnishing
    -0.07
    efined
    -0.07
     tilante
    -0.07
     photographer
    -0.07
     glacier
    -0.07
    landse
    -0.07
    POSITIVE LOGITS
     ignore
    0.12
    Ignore
    0.11
     aside
    0.11
    0.11
     Ignore
    0.10
    _ignore
    0.10
     irrelevant
    0.10
     ignor
    0.10
    ignored
    0.09
     notwithstanding
    0.09
    Act Density 0.054%

    No Known Activations