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

    Neuronpedia

    INDEX
    Explanations

    overview

    np_max-act · gemini-2.0-flash

    The neuron specifically fires on section-heading tokens—most notably the word “Overview.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Investment
    -0.07
    _into
    -0.07
    DETAIL
    -0.06
    acho
    -0.06
    Alex
    -0.06
     Daw
    -0.06
    	random
    -0.06
    AINS
    -0.06
    013
    -0.06
     نو
    -0.06
    POSITIVE LOGITS
    елеф
    0.06
    칙
    0.06
     Guide
    0.06
     pale
    0.06
    roids
    0.06
     Specialist
    0.06
    _LEFT
    0.06
     ark
    0.06
    attachments
    0.05
     lombok
    0.05
    Act Density 0.007%

    No Known Activations