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

    Neuronpedia

    INDEX
    Explanations

    historical events

    np_max-act · gemini-2.0-flash

    The neuron primarily detects numeric tokens corresponding to years or dates in historical text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    україн
    -0.07
    madığı
    -0.07
     �
    -0.07
     filament
    -0.07
    046
    -0.06
    	BYTE
    -0.06
    amu
    -0.06
     bw
    -0.06
    ilmiş
    -0.06
    [MAXN
    -0.06
    POSITIVE LOGITS
     Scott
    0.07
     cannon
    0.07
     خارجية
    0.06
    的手
    0.06
     _
    0.06
    theon
    0.06
    {}{↵
    0.06
     surprises
    0.06
     Street
    0.06
     \
    0.06
    Act Density 0.013%

    No Known Activations