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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    The neuron flags tokens that are method or function names—especially common IO or formatting calls like “format,” “read,” “write,” and similar invoked methods.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Iso
    -0.07
     Redistribution
    -0.06
     OTHERWISE
    -0.06
     drone
    -0.06
    への
    -0.06
    들을
    -0.06
    376
    -0.06
    .newBuilder
    -0.06
    ıza
    -0.06
     speedy
    -0.06
    POSITIVE LOGITS
    acket
    0.07
    MPI
    0.06
     Paula
    0.06
    คาส
    0.06
     Killer
    0.06
    سی
    0.06
    kdir
    0.06
     Luc
    0.06
    rotein
    0.06
     commentaire
    0.06
    Act Density 0.068%

    No Known Activations