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

    Neuronpedia

    INDEX
    Explanations

    comparisons

    np_max-act · gemini-2.0-flash

    This neuron detects numeric tokens containing decimal points (the fractional part of numbers).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ornado
    -0.07
     кри
    -0.07
    ISIBLE
    -0.06
     �
    -0.06
    asures
    -0.06
     lunch
    -0.06
    ρία
    -0.06
    егист
    -0.06
    >alert
    -0.06
    ucus
    -0.06
    POSITIVE LOGITS
     lavor
    0.07
     geliyor
    0.06
    isk
    0.06
     Elekt
    0.06
     castle
    0.06
     prix
    0.06
    改
    0.06
    "))
↵
    0.06
    ２
    0.06
    staw
    0.06
    Act Density 0.015%

    No Known Activations