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

    Neuronpedia

    INDEX
    Explanations

    classify

    np_max-act · gemini-2.0-flash

    The neuron fires on structural markup tokens—i.e. angle-bracketed tags and special “<…>” start/end markers rather than natural language words.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _arguments
    -0.06
     Однако
    -0.06
    ош
    -0.06
     lacked
    -0.06
     nonatomic
    -0.06
     WaitFor
    -0.06
     Fairy
    -0.06
    Dispose
    -0.06
    oš
    -0.06
    SignIn
    -0.06
    POSITIVE LOGITS
    かの
    0.07
    -Re
    0.06
    .var
    0.06
    BY
    0.06
    'S
    0.06
    .drawString
    0.06
    ตร
    0.06
    根本
    0.06
     sollten
    0.06
    -tra
    0.06
    Act Density 0.000%

    No Known Activations