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

    Neuronpedia

    INDEX
    Explanations

    of, in

    np_max-act · gemini-2.0-flash

    This neuron selectively activates on high-frequency short function words—especially prepositions and articles like “of,” “in,” and “the.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Esper
    -0.06
     Temple
    -0.06
     pasa
    -0.06
     Emoji
    -0.06
     Painter
    -0.06
    Watching
    -0.06
    veis
    -0.06
    fleet
    -0.06
    Parent
    -0.06
    Century
    -0.06
    POSITIVE LOGITS
     کسب
    0.07
     گردید
    0.07
     magg
    0.07
    ระบ
    0.07
    ’une
    0.07
    .ShowDialog
    0.06
     cerv
    0.06
    'une
    0.06
     #+#
    0.06
    (annotation
    0.06
    Act Density 0.102%

    No Known Activations