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

    Neuronpedia

    INDEX
    Explanations

    numbers

    np_max-act · gemini-2.0-flash

    The neuron activates on the numeral that specifies how many items to list (e.g. the “5” in “list me 5 …”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     foul
    -0.07
    	play
    -0.06
     поп
    -0.06
     Array
    -0.06
    112
    -0.06
     engine
    -0.06
    113
    -0.06
     costs
    -0.06
     vaccine
    -0.06
     soils
    -0.06
    POSITIVE LOGITS
    zas
    0.08
     pornofil
    0.07
    ADING
    0.07
    haus
    0.07
    χές
    0.07
    ोषण
    0.06
    defs
    0.06
    cen
    0.06
    iddi
    0.06
     счита
    0.06
    Act Density 0.022%

    No Known Activations