INDEX
    Explanations

    Statistical analysis

    The neuron is sensitive to occurrences of the definite article “The” (especially when starting key sentences in methods/results), and more broadly to sentence-initial discourse markers like “The,” “A,” “Moreover,” “Univariate,” or “Multivariate.”

    New Auto-Interp
    Negative Logits
    footer
    -0.07
    hatt
    -0.06
    .Properties
    -0.06
    -mf
    -0.06
    _reports
    -0.06
     Johns
    -0.06
     상품
    -0.06
     Barnes
    -0.06
    Inline
    -0.06
     adap
    -0.06
    POSITIVE LOGITS
    PRESENT
    0.07
    ،↵
    0.07
    еку
    0.07
     MIC
    0.07
     circ
    0.07
    cion
    0.06
    odynamic
    0.06
     incess
    0.06
    ctica
    0.06
    {})
    0.06
    Act Density 0.049%

    No Known Activations