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

    Neuronpedia

    INDEX
    Explanations

    names

    np_max-act · gemini-2.0-flash

    The neuron is detecting occurrences of identifiers related to filenames in C code (e.g. “filename,” “filename_len,” string parameters for file‐handling functions).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    atum
    -0.07
    ула
    -0.06
    hest
    -0.06
    ança
    -0.06
    幾
    -0.06
     Зап
    -0.06
     layoffs
    -0.06
    ']]
    -0.06
    "<
    -0.06
    'We
    -0.06
    POSITIVE LOGITS
    (dot
    0.07
     MOT
    0.07
    Char
    0.07
    []{
    0.07
    _ut
    0.07
    /terms
    0.06
    .USER
    0.06
     Wat
    0.06
     Beaut
    0.06
     erot
    0.06
    Act Density 0.083%

    No Known Activations