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

    Neuronpedia

    INDEX
    Explanations

    FLOPS and computer performance

    np_max-act · gemini-2.0-flash

    This neuron detects numeric specification tokens—especially performance numbers and units like FLOPS, memory sizes, and transistor counts—in technical hardware descriptions.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ushort
    -0.07
     CONTRIBUTORS
    -0.07
    K
    -0.07
     Profession
    -0.06
     Han
    -0.06
    (us
    -0.06
    mak
    -0.06
    Professional
    -0.06
    usuario
    -0.06
    ")↵↵
    -0.06
    POSITIVE LOGITS
    经营
    0.07
     quo
    0.06
    ая
    0.06
    Portfolio
    0.06
    /firebase
    0.06
    ضو
    0.06
    _ENSURE
    0.06
     ethnic
    0.06
     코로나
    0.05
     нами
    0.05
    Act Density 0.010%

    No Known Activations