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

    Neuronpedia

    INDEX
    Explanations

    [@

    np_max-act · gemini-2.0-flash

    This neuron activates on academic citation markers (the bracketed reference tokens like “[@…]”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     vyrob
    -0.07
     questi
    -0.07
     میان
    -0.06
    unnel
    -0.06
     trưởng
    -0.06
    .Cmd
    -0.06
    งม
    -0.06
     спроб
    -0.06
     suspected
    -0.06
     premi
    -0.06
    POSITIVE LOGITS
    (column
    0.07
     overflowing
    0.06
    Happy
    0.06
    (matrix
    0.06
    (query
    0.06
     Quaternion
    0.06
     Fauc
    0.06
     الف
    0.06
     [],↵
    0.06
    query
    0.06
    Act Density 0.009%

    No Known Activations