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

    Neuronpedia

    INDEX
    Explanations

    1900s years

    np_max-act · gemini-2.0-flash

    This neuron selectively activates on numeric tokens, especially years and numeric citation details.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _CHUNK
    -0.07
    _lines
    -0.07
    -data
    -0.06
    .tmp
    -0.06
     چیز
    -0.06
     Hyderabad
    -0.06
     >↵↵
    -0.06
    یشه
    -0.06
     seq
    -0.06
    <View
    -0.06
    POSITIVE LOGITS
    hus
    0.06
     horrend
    0.06
     locom
    0.06
     disponible
    0.06
     Ав
    0.06
    DOWN
    0.06
    承
    0.06
     affirm
    0.06
     AF
    0.06
    也不
    0.06
    Act Density 0.009%

    No Known Activations