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

    Neuronpedia

    INDEX
    Explanations

    Retrieving something

    np_max-act · gemini-2.0-flash

    commands or requests related to objects and their retrieval.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron activates on imperative requests asking someone to retrieve or deliver a specified object.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     traged
    -0.08
    智
    -0.07
    _rule
    -0.07
    lığı
    -0.07
    _wall
    -0.06
    Way
    -0.06
     SSP
    -0.06
     spi
    -0.06
     shar
    -0.06
    _statistics
    -0.06
    POSITIVE LOGITS
    mith
    0.07
    <F
    0.07
     description
    0.07
    .Ş
    0.06
     retrieve
    0.06
     retrieving
    0.06
     hiçbir
    0.06
    无码
    0.06
    ainted
    0.06
    明
    0.06
    Act Density 0.014%

    No Known Activations