INDEX
    Explanations

    The neuron responds to thematic section-header vocabulary—labels or descriptors (e.g. “national,” “global,” “security,” “terrorism,” “intelligence”) that mark a news section or category.

    New Auto-Interp
    Negative Logits
    -0.07
     فنی
    -0.07
    (@"%@",
    -0.07
    ่วน
    -0.06
    Showing
    -0.06
    -0.06
    	bs
    -0.06
    'http
    -0.06
    olas
    -0.06
     //----------------------------------------------------------------
    -0.06
    POSITIVE LOGITS
     Chair
    0.07
    .Builder
    0.07
     Cult
    0.06
     Сам
    0.06
    -preview
    0.06
    ,Th
    0.06
     Strike
    0.06
     planes
    0.06
     Gibraltar
    0.06
     intéress
    0.06
    Act Density 0.028%

    No Known Activations