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

    Neuronpedia

    INDEX
    Explanations

    computer science

    np_max-act · gemini-2.0-flash

    The neuron activates on meta-references to the content being processed (words like “text,” “script,” “information,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    istrib
    -0.07
    .large
    -0.06
     Terminal
    -0.06
     currents
    -0.06
    Absolute
    -0.06
     monkeys
    -0.06
    رده
    -0.06
    -layer
    -0.06
     lawyer
    -0.06
    .bytes
    -0.06
    POSITIVE LOGITS
    .tc
    0.07
     Tout
    0.07
    viewport
    0.07
    instances
    0.07
    esser
    0.06
     Lucifer
    0.06
    fü
    0.06
    CTX
    0.06
     ferment
    0.06
     ZX
    0.06
    Act Density 0.044%

    No Known Activations