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

    Neuronpedia

    INDEX
    Explanations

    Code

    np_max-act · gemini-2.0-flash

    The neuron specifically fires on lines that update one of the guess‐range bounds (e.g. “high_num = number;” or “low_num = number;”) in the switch‐case.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     cz
    -0.08
    .Rectangle
    -0.07
    condition
    -0.07
    ↵				↵
    -0.07
     Genç
    -0.06
    ↵		↵
    -0.06
    ┃
    -0.06
     Gujarat
    -0.06
    Những
    -0.06
    mongoose
    -0.06
    POSITIVE LOGITS
    保证
    0.06
     служби
    0.06
    _COMPONENT
    0.06
     nomin
    0.06
    _fe
    0.06
    رت
    0.06
     quilt
    0.06
    يلاد
    0.06
    'є
    0.06
    爷
    0.06
    Act Density 0.038%

    No Known Activations