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

    Neuronpedia

    INDEX
    Explanations

    8

    np_max-act · gemini-2.0-flash

    The neuron detects date stamps—particularly the numeric year (and associated day/month) elements in article headers.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    adamente
    -0.07
    σεων
    -0.06
     Bangladesh
    -0.06
    ￣￣￣￣￣￣￣￣
    -0.06
     gerekmektedir
    -0.06
    Intent
    -0.06
    ｀ヽ
    -0.06
    zia
    -0.06
    二十
    -0.06
     tran
    -0.06
    POSITIVE LOGITS
    _DEPEND
    0.06
     Simone
    0.06
     Пр
    0.06
     ric
    0.06
    fires
    0.06
    .friends
    0.06
    Drawing
    0.06
     Dave
    0.06
     đàn
    0.06
    natural
    0.06
    Act Density 0.011%

    No Known Activations