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

    Neuronpedia

    INDEX
    Explanations

    ih

    np_max-act · gemini-2.0-flash

    The neuron is specialized for recognizing fragments of BitTorrent “magnet:” URIs (e.g. “magnet:?xt=urn:btih:…”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Liberation
    -0.06
    !--
    -0.06
    {i
    -0.06
     Wol
    -0.06
    .Bl
    -0.06
    .scale
    -0.06
    Im
    -0.06
     علی
    -0.06
    Mes
    -0.06
    (dataset
    -0.06
    POSITIVE LOGITS
    をする
    0.07
    нообраз
    0.07
    .drop
    0.07
     ranks
    0.07
    .shortcuts
    0.07
    ковод
    0.06
    ayi
    0.06
     shapes
    0.06
     �
    0.06
     Yates
    0.06
    Act Density 0.001%

    No Known Activations