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

    Neuronpedia

    INDEX
    Explanations

    proper nouns, capitalized words

    np_max-act · gemini-2.0-flash

    The neuron reliably detects mentions of “federated learning.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .pow
    -0.07
    .Π
    -0.06
    ialog
    -0.06
    	while
    -0.06
    panies
    -0.06
     Нат
    -0.06
    ccount
    -0.06
    .Split
    -0.06
    .addr
    -0.06
    çu
    -0.06
    POSITIVE LOGITS
     ang
    0.07
    =R
    0.07
    Ф
    0.07
    189
    0.07
     arreglo
    0.06
    )'),
    0.06
     hamburger
    0.06
    uffed
    0.06
    >>(↵
    0.06
    “,
    0.06
    Act Density 0.051%

    No Known Activations