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

    Neuronpedia

    INDEX
    Explanations

    Controller

    np_max-act · gemini-2.0-flash

    The neuron fires on Java exception stack‐trace frames—i.e. fully qualified class and method names (like java.security.AccessController.doPrivileged) in “at …” lines.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     محل
    -0.06
    реш
    -0.06
     رسم
    -0.06
    empt
    -0.06
     tổn
    -0.06
    許
    -0.06
     massac
    -0.06
    agged
    -0.06
     suede
    -0.06
    驶
    -0.06
    POSITIVE LOGITS
    hur
    0.07
    ACTER
    0.07
     belongings
    0.07
    091
    0.06
     Dustin
    0.06
    lords
    0.06
     entertain
    0.06
     class
    0.06
     brain
    0.06
    ând
    0.06
    Act Density 0.001%

    No Known Activations