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

    Neuronpedia

    INDEX
    Explanations

    experiments, measurements

    np_max-act · gemini-2.0-flash

    The neuron fires on tokens that describe experimental methods and procedural steps (e.g. “measured,” “performed,” “investigated,” “assessed”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .parse
    -0.07
    }))↵
    -0.07
     cruiser
    -0.06
    инку
    -0.06
    benchmark
    -0.06
    õ
    -0.06
    kat
    -0.06
     CPS
    -0.06
     IMF
    -0.06
    ونت
    -0.06
    POSITIVE LOGITS
     demok
    0.07
    tsy
    0.07
     Docker
    0.07
     nodded
    0.07
    да
    0.06
     знаком
    0.06
    (),
↵
    0.06
    ld
    0.06
     слов
    0.06
    _Link
    0.06
    Act Density 0.085%

    No Known Activations