INDEX
    Explanations

    captions related to news articles or images

    captions that indicate a mourning or memorial theme

    New Auto-Interp
    Negative Logits
    ught
    -0.96
    abol
    -0.87
    bye
    -0.74
    axis
    -0.74
    sed
    -0.73
    omics
    -0.70
    neg
    -0.69
    ult
    -0.69
    tein
    -0.66
    nesia
    -0.64
    POSITIVE LOGITS
     Thumbnails
    1.21
     Caption
    1.20
     Close
    0.98
     Loading
    0.89
     largeDownload
    0.85
     Shooting
    0.85
     ï
    0.85
     Scenes
    0.82
     Execution
    0.79
     Continued
    0.79
    Act Density 0.004%

    No Known Activations