INDEX
    Explanations

    references to web engagement and community contributions

    New Auto-Interp
    Negative Logits
    Batch
    -0.16
     Baum
    -0.16
     Bracket
    -0.15
    hana
    -0.14
     Discord
    -0.14
     Bias
    -0.14
    Brief
    -0.14
     Backbone
    -0.14
    discord
    -0.14
     Batch
    -0.14
    POSITIVE LOGITS
     blog
    0.71
    blog
    0.66
     Blog
    0.65
    -blog
    0.61
     blogs
    0.61
    Blog
    0.60
     blogging
    0.59
    _blog
    0.54
     bloggers
    0.53
    .blog
    0.50
    Act Density 0.206%

    No Known Activations