INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    صيل
    -0.08
    Jes
    -0.08
    YO
    -0.08
    മേ
    -0.07
    Centre
    -0.07
     Terrace
    -0.07
    -Marie
    -0.07
     Breakdown
    -0.07
    Center
    -0.07
    .Write
    -0.07
    POSITIVE LOGITS
     fake
    0.13
    .fake
    0.12
     pretending
    0.12
    假的
    0.12
     Fake
    0.12
     faux
    0.12
     pretend
    0.11
    fake
    0.11
    (fake
    0.11
     artificially
    0.11
    Act Density 0.094%

    No Known Activations