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

    Neuronpedia

    INDEX
    Explanations

    Assistant

    np_max-act · gemini-2.0-flash

    the start of an assistant’s message in chat-style formatting (the assistant turn boundary).

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Negative Logits
    CAP
    -0.08
     труд
    -0.07
    cec
    -0.07
    厂房
    -0.07
    彩
    -0.07
     Gaza
    -0.07
    苑
    -0.07
    hyper
    -0.07
    etre
    -0.06
    Pipeline
    -0.06
    POSITIVE LOGITS
     thaimassage
    0.07
    睎
    0.07
    _seqs
    0.07
    _FILES
    0.07
     distancia
    0.07
    .IO
    0.07
    升华
    0.06
    뽁
    0.06
    QUOTE
    0.06
    ()`
    0.06
    Act Density 0.106%

    No Known Activations