INDEX
    Explanations

    expressions of sympathy or concern regarding loss and support for humanitarian efforts

    New Auto-Interp
    Negative Logits
     cheap
    -0.53
    thard
    -0.50
    cheap
    -0.48
     goodies
    -0.48
     busted
    -0.48
    InjectMocks
    -0.47
     big
    -0.47
    يع
    -0.46
     screamed
    -0.46
     UIKit
    -0.46
    POSITIVE LOGITS
     pleaſure
    0.77
     neceſſ
    0.77
     Roskov
    0.76
     raiſ
    0.74
     itſelf
    0.73
     fubject
    0.72
     myſelf
    0.71
    tanleria
    0.71
     poffe
    0.71
    ImageContext
    0.70
    Act Density 0.319%

    No Known Activations