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

    Neuronpedia

    INDEX
    Explanations

    Legal cases

    np_max-act · gemini-2.0-flash

    This neuron detects bracketed legal or scholarly reference markers (numeric citations or footnote indicators).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    >{"
    -0.07
    .arm
    -0.07
    ोन
    -0.06
     输入
    -0.06
    .format
    -0.06
     эксп
    -0.06
    -all
    -0.06
     GT
    -0.06
    ContextMenu
    -0.06
    }/>
    -0.06
    POSITIVE LOGITS
    에는
    0.07
    lew
    0.06
    がい
    0.06
    альными
    0.06
    ٤
    0.06
     podporu
    0.06
     pob
    0.06
     اما
    0.06
     olası
    0.06
     niż
    0.06
    Act Density 0.003%

    No Known Activations