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

    Neuronpedia

    INDEX
    Explanations

    untreated

    np_max-act · gemini-2.0-flash

    This neuron is specifically triggered by the word “untreated” (marking mentions of untreated conditions).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Animator
    -0.07
    ){↵↵
    -0.07
     голов
    -0.06
    (listener
    -0.06
    .onclick
    -0.06
    	Log
    -0.06
    еріга
    -0.06
    [random
    -0.06
    ۳۶
    -0.06
     самое
    -0.06
    POSITIVE LOGITS
     untreated
    0.07
    rence
    0.06
     athletic
    0.06
    _shell
    0.06
     Offices
    0.06
    ий
    0.06
    osterone
    0.06
     pylab
    0.06
    _Server
    0.06
     behaved
    0.06
    Act Density 0.009%

    No Known Activations