INDEX
    Explanations

    publications

    This neuron fires on tokens that are part of titles or names of creative works and artists (e.g. book titles, series names, band or album names).

    New Auto-Interp
    Negative Logits
     sir
    -0.07
    λος
    -0.06
     declining
    -0.06
    .High
    -0.06
     Fairfax
    -0.06
    ersion
    -0.06
    ۵۰
    -0.06
     package
    -0.06
    จำ
    -0.06
    Hi
    -0.06
    POSITIVE LOGITS
    (datos
    0.07
     siguientes
    0.06
    美元
    0.06
     коп
    0.06
    (selected
    0.06
     parametros
    0.06
     конкрет
    0.06
    Filtered
    0.06
    icros
    0.06
     eliminar
    0.06
    Act Density 0.110%

    No Known Activations