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

    Neuronpedia

    INDEX
    Explanations

    .read

    np_max-act · gemini-2.0-flash

    The neuron fires on code patterns that read or split file contents—i.e. calls like `.read(...)` or `.split(...)` used to load text from a file.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     приводит
    -0.06
    _curve
    -0.06
     законодав
    -0.06
    icipation
    -0.06
    Download
    -0.06
    assic
    -0.06
    	ArrayList
    -0.06
    orative
    -0.06
     ayrıntılı
    -0.06
    Women
    -0.06
    POSITIVE LOGITS
     ifdef
    0.07
    STYLE
    0.07
    ブロ
    0.06
    AMA
    0.06
     chod
    0.06
    рог
    0.06
     instability
    0.06
    마
    0.06
     Proceedings
    0.06
    enor
    0.06
    Act Density 0.003%

    No Known Activations