Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    (

    np_max-act · gemini-2.0-flash

    This neuron fires on the parenthetical markers of news‐article datelines—especially the “(AP)”-style or station‐ID parentheses and adjacent city/state abbreviations that appear at the top of wire stories.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     kennen
    -0.07
    оро
    -0.06
    olvable
    -0.06
    奶
    -0.06
     Theo
    -0.06
    lanmış
    -0.06
    (valor
    -0.06
     architekt
    -0.06
    .Row
    -0.06
     rij
    -0.06
    POSITIVE LOGITS
     <-
    0.07
    big
    0.07
    _PREF
    0.06
     рез
    0.06
    .people
    0.06
     cif
    0.06
     Accent
    0.06
     device
    0.06
    ás
    0.06
     Суд
    0.06
    Act Density 0.003%

    No Known Activations