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

    Neuronpedia

    INDEX
    Explanations

    algebraic expressions

    np_max-act · gemini-2.0-flash

    This neuron detects algebraic terms expressing a variable raised to a power (e.g. “*x**2”–style exponent notation).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    bulan
    -0.07
     Florian
    -0.06
    ("/:
    -0.06
    查询
    -0.06
    _circle
    -0.06
     "",↵
    -0.06
     embassy
    -0.06
     Browser
    -0.06
    istogram
    -0.06
    iate
    -0.06
    POSITIVE LOGITS
     Sugar
    0.06
     JNI
    0.06
    0.06
     cookies
    0.06
    AF
    0.06
    다
    0.06
    lj
    0.06
     bake
    0.06
    0.06
    .Real
    0.06
    Act Density 0.008%

    No Known Activations