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

    Neuronpedia

    INDEX
    Explanations

    File

    np_max-act · gemini-2.0-flash

    The neuron fires on ordinary English words appearing in code comments or explanatory prose rather than on actual code syntax.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     así
    -0.07
     daha
    -0.07
     Luật
    -0.07
    argc
    -0.07
    ẩn
    -0.07
    usher
    -0.06
     freaking
    -0.06
    undy
    -0.06
    ３
    -0.06
    拟
    -0.06
    POSITIVE LOGITS
    /exp
    0.07
     الد
    0.07
     Email
    0.06
    _U
    0.06
    -fix
    0.06
    ニメ
    0.06
     BO
    0.06
    (In
    0.06
     pcb
    0.06
     शहर
    0.06
    Act Density 0.031%

    No Known Activations