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

    Neuronpedia

    INDEX
    Explanations

    Parts

    np_max-act · gemini-2.0-flash

    The neuron fires on square‐bracket index operations (e.g. accessing elements via var[...]).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    unlikely
    -0.07
     mig
    -0.06
    _Global
    -0.06
     Defensive
    -0.06
    PrototypeOf
    -0.06
    _visibility
    -0.06
     taxis
    -0.06
    Crypto
    -0.06
    「お
    -0.06
     cpu
    -0.06
    POSITIVE LOGITS
    [curr
    0.07
    [child
    0.06
    instanc
    0.06
     статус
    0.06
     Situation
    0.06
    ере
    0.06
     OPER
    0.06
    アメリカ
    0.06
     Entrance
    0.06
    ERENCE
    0.06
    Act Density 0.051%

    No Known Activations