INDEX
    Explanations

    Horizon and horizontal contexts

    New Auto-Interp
    Negative Logits
     horror
    0.58
     hören
    0.55
     हारे
    0.51
     halaman
    0.48
     heroic
    0.48
     horrors
    0.47
     harem
    0.47
     hemorrhage
    0.46
     hearing
    0.46
     hydrolysis
    0.46
    POSITIVE LOGITS
    ंगाबाद
    0.52
    থা
    0.48
    izontally
    0.47
    ontal
    0.47
    షి
    0.44
    வு
    0.42
    0.42
     }^{+}
    0.41
    тали
    0.40
    ثم
    0.39
    Act Density 0.007%

    No Known Activations