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

    Neuronpedia

    INDEX
    Explanations

    ");"

    np_max-act · gemini-2.0-flash

    The neuron activates on the string literal pattern License"); in source‐code comments, i.e. it detects occurrences of that exact license reference token.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    /read
    -0.08
    .Ad
    -0.07
     airport
    -0.07
    hots
    -0.07
     jet
    -0.07
     d
    -0.07
    :'
    -0.06
     fo
    -0.06
    Beans
    -0.06
     :↵
    -0.06
    POSITIVE LOGITS
    ");
    0.07
     станет
    0.07
    ());
    0.06
     korun
    0.06
     Bylo
    0.06
    斯特
    0.06
     ballots
    0.06
    wg
    0.06
     그렇
    0.06
     والتي
    0.06
    Act Density 0.000%

    No Known Activations