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

    Neuronpedia

    INDEX
    Explanations

    j

    np_max-act · gemini-2.0-flash

    The neuron fires on JNI interop identifiers (tokens like JNIEnv, jclass, jmethodID, jobject, jbyteArray, etc.) indicating native‐to‐Java interface code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     rk
    -0.07
    �
    -0.06
    (BigDecimal
    -0.06
     betrayal
    -0.06
    swers
    -0.06
    -clock
    -0.06
    럴
    -0.06
     opatření
    -0.06
    <W
    -0.06
    (grammarAccess
    -0.06
    POSITIVE LOGITS
    etine
    0.06
    imizer
    0.06
     skim
    0.06
    ุ
    0.06
    orderId
    0.06
    ☴
    0.06
     Swim
    0.06
    hib
    0.06
    40
    0.06
    CGFloat
    0.06
    Act Density 0.003%

    No Known Activations