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

    Neuronpedia

    INDEX
    Explanations

    Time periods and breaks

    np_max-act · gemini-2.0-flash

    The neuron activates on references to dates and time spans, especially year numbers and words like “years” or “to.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    iner
    -0.07
    _anim
    -0.07
     lawmakers
    -0.06
     modes
    -0.06
     patrol
    -0.06
    002
    -0.06
    (hWnd
    -0.06
     scrap
    -0.06
    	conf
    -0.06
    YELLOW
    -0.06
    POSITIVE LOGITS
     vý
    0.07
     fearless
    0.07
    робіт
    0.07
     الدر
    0.07
    phies
    0.07
     tây
    0.06
    -Isl
    0.06
     Ö
    0.06
    oce
    0.06
     Representation
    0.06
    Act Density 0.021%

    No Known Activations