INDEX
    Explanations

    brevity and conciseness

    New Auto-Interp
    Negative Logits
    picable
    0.43
    nous
    0.42
     derogatory
    0.41
    security
    0.40
     security
    0.40
    gebras
    0.39
    岡山
    0.39
     pathogenic
    0.39
    wijk
    0.38
    exists
    0.38
    POSITIVE LOGITS
     shorter
    0.98
     brevity
    0.91
     корот
    0.84
    0.80
     shortened
    0.78
     shorten
    0.76
     concise
    0.75
    0.75
     kısa
    0.74
     مختصر
    0.73
    Act Density 0.149%

    No Known Activations