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

    Neuronpedia

    INDEX
    Explanations

    blog-like posts

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on culinary ingredient words (e.g. “olive,” “vinegar,” “salt,” “pepper”) in recipe‐style text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    perm
    -0.07
    .Fields
    -0.06
     Rath
    -0.06
     Erik
    -0.06
    defines
    -0.06
     rinse
    -0.06
    ()]
    -0.06
    112
    -0.06
    694
    -0.05
     Drink
    -0.05
    POSITIVE LOGITS
    umbs
    0.07
     blackjack
    0.07
    르고
    0.07
    (equalTo
    0.07
     navegador
    0.07
     глав
    0.06
     крок
    0.06
     stream
    0.06
    bol
    0.06
     InterruptedException
    0.06
    Act Density 0.208%

    No Known Activations