INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     bored
    0.54
     annoyed
    0.47
    weird
    0.46
     weird
    0.46
     Centuries
    0.44
     Tired
    0.43
     assortment
    0.43
     boredom
    0.42
    rusty
    0.42
     disgusted
    0.42
    POSITIVE LOGITS
     tragically
    1.48
     beloved
    1.40
     deceased
    1.33
     loved
    1.32
     tragic
    1.27
     tragedy
    1.19
     погиб
    1.18
    1.16
     Trag
    1.13
    去世
    1.09
    Act Density 0.042%

    No Known Activations