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

    Neuronpedia

    INDEX
    Explanations

    URLs and social media

    np_max-act · gemini-2.0-flash

    The neuron detects substrings of URLs and social‐media handles (i.e. domain names, path or handle fragments in web links).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     stumbled
    -0.07
     enthusiasts
    -0.06
    음을
    -0.06
    encias
    -0.06
    ObjectId
    -0.06
     vede
    -0.06
    osome
    -0.06
     McDonald
    -0.06
    oggled
    -0.06
     calves
    -0.06
    POSITIVE LOGITS
    頁
    0.06
    _scheme
    0.06
    ，由
    0.06
     picks
    0.06
    changing
    0.06
    setup
    0.06
     страш
    0.06
     disag
    0.06
     concess
    0.06
     additionally
    0.06
    Act Density 0.011%

    No Known Activations