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

    Neuronpedia

    INDEX
    Explanations

    So

    np_max-act · gemini-2.0-flash

    The neuron fires on introductory or transition tokens that launch instructional or step‐by‐step guidance—especially the components of “To get started,” at the beginning of a how-to section.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     sider
    -0.07
     추가
    -0.07
    ปร
    -0.06
    털
    -0.06
     penetrating
    -0.06
     yoksa
    -0.06
     호
    -0.06
     Terrace
    -0.06
    rz
    -0.06
    FN
    -0.06
    POSITIVE LOGITS
    	EIF
    0.07
    Brain
    0.07
    .selector
    0.06
    .lesson
    0.06
    Fuel
    0.06
     pricing
    0.06
    game
    0.06
     Heal
    0.06
     grip
    0.06
     tram
    0.06
    Act Density 0.081%

    No Known Activations