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

    Neuronpedia

    INDEX
    Explanations

    Scientific Errors

    np_max-act · gemini-2.0-flash

    The neuron activates on language typical of stating scientific results or predictions—words like “produce,” “correct,” “predict,” “spurious,” “transition,” and “forbidden” that describe theoretical findings or their validation.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    igated
    -0.08
    онт
    -0.06
    ु�
    -0.06
     activated
    -0.06
    循
    -0.06
     affine
    -0.06
     chip
    -0.06
    ishops
    -0.06
    -cycle
    -0.06
    ug
    -0.06
    POSITIVE LOGITS
     Output
    0.07
     Describe
    0.07
     pname
    0.07
     Pere
    0.07
     Beard
    0.06
    [,
    0.06
    ErrorException
    0.06
    /{{
    0.06
     бар
    0.06
     Statements
    0.06
    Act Density 0.031%

    No Known Activations