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

    Neuronpedia

    INDEX
    Explanations

    May

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on proper nouns or named entities (e.g. names of people, places, or organizations).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _checks
    -0.06
     civilized
    -0.06
     write
    -0.06
    Reuters
    -0.06
    ;-
    -0.06
    .Cloud
    -0.06
     marijuana
    -0.06
    HttpException
    -0.06
    $current
    -0.05
     BitConverter
    -0.05
    POSITIVE LOGITS
    (dynamic
    0.09
    алися
    0.07
    orde
    0.07
    PreferredSize
    0.07
     GAL
    0.07
    ORIZ
    0.07
    acobian
    0.07
    ционных
    0.06
    ález
    0.06
    ^{-
    0.06
    Act Density 0.523%

    No Known Activations