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

    Neuronpedia

    INDEX
    Explanations

    Brace

    np_max-act · gemini-2.0-flash

    The neuron primarily activates on numeric literals in code, identifying tokens that represent numbers.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ��
    -0.08
     differentiate
    -0.07
     receipt
    -0.07
    “In
    -0.07
    	ref
    -0.06
    ibrate
    -0.06
    developer
    -0.06
    Bug
    -0.06
    /logger
    -0.06
     surrounds
    -0.06
    POSITIVE LOGITS
     POR
    0.07
     χρό
    0.07
    kah
    0.07
    (image
    0.06
    setFont
    0.06
    _PIN
    0.06
     дев
    0.06
    FieldValue
    0.06
    _mr
    0.06
    rd
    0.06
    Act Density 0.004%

    No Known Activations