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

    Neuronpedia

    INDEX
    Explanations

    Number of items to purchase

    np_max-act · gemini-2.0-flash

    The neuron flags instructions specifying numeric quantities (e.g. “number of additional users/licenses,” “number of copies,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     positives
    -0.07
    ceeded
    -0.07
     fore
    -0.06
     partir
    -0.06
     accessor
    -0.06
     padx
    -0.06
     surgeries
    -0.06
     trp
    -0.06
    vekili
    -0.06
     dij
    -0.06
    POSITIVE LOGITS
    Replacement
    0.07
    (JSON
    0.06
    owns
    0.06
     accountability
    0.06
    355
    0.06
    maktan
    0.06
    пп
    0.06
     Instruments
    0.06
    _active
    0.06
    /e
    0.06
    Act Density 0.001%

    No Known Activations