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

    Neuronpedia

    INDEX
    Explanations

    square

    np_max-act · gemini-2.0-flash

    This neuron detects numeric quantity expressions, especially measurements (e.g., numbers followed by units like “square meters”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    -worthy
    -0.07
    periences
    -0.07
    ७
    -0.07
    _settings
    -0.07
    Objects
    -0.07
    _rsa
    -0.07
    (agent
    -0.07
    ertime
    -0.06
    _speed
    -0.06
    WASHINGTON
    -0.06
    POSITIVE LOGITS
     ак
    0.06
     Tyr
    0.06
     sitio
    0.06
    lk
    0.06
    .slim
    0.06
    \",↵
    0.06
     provinc
    0.06
     pb
    0.06
     conserv
    0.06
     trúc
    0.06
    Act Density 0.003%

    No Known Activations