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

    Neuronpedia

    INDEX
    Explanations

    )arg

    np_max-act · gemini-2.0-flash

    This neuron consistently activates on the numeric “argN” placeholders (the argument‐index tokens) in Objective-C method signatures.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _WAIT
    -0.07
     perception
    -0.07
     pleasures
    -0.07
     noses
    -0.07
     dul
    -0.07
    iot
    -0.07
     DAT
    -0.06
     paci
    -0.06
    府
    -0.06
     viewers
    -0.06
    POSITIVE LOGITS
    )arg
    0.08
     ),
↵
    0.07
    プラ
    0.07
     représ
    0.07
    ++){
↵
    0.06
    /**↵
    0.06
    '):
↵
    0.06
    ۱۹۶
    0.06
    ら
    0.06
    ]){
↵
    0.06
    Act Density 0.001%

    No Known Activations