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

    Neuronpedia

    INDEX
    Explanations

    25

    np_max-act · gemini-2.0-flash

    This neuron detects numeric score values (e.g. set or match scores like “25-17,” “25-21,” etc.) in sports reporting.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    挂
    -0.07
    	md
    -0.06
    扎
    -0.06
    agement
    -0.06
    modity
    -0.06
    φέ
    -0.06
     attaches
    -0.06
    598
    -0.06
     �
    -0.06
    @brief
    -0.05
    POSITIVE LOGITS
    robot
    0.07
     (){↵
    0.07
    .volley
    0.07
     remotely
    0.07
    ór
    0.07
    (handler
    0.07
     misuse
    0.07
    ी.
    0.07
    KN
    0.06
     Tamil
    0.06
    Act Density 0.077%

    No Known Activations