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

    Neuronpedia

    INDEX
    Explanations

    What

    np_max-act · gemini-2.0-flash

    This neuron fires on question or instruction prompts—words like “Find,” “Calculate,” “What,” or “List” that introduce a problem statement.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    StringBuilder
    -0.07
    cxx
    -0.07
    bos
    -0.07
    ');
↵
↵
    -0.06
     broadcasting
    -0.06
     Gala
    -0.06
    ยอด
    -0.06
     Three
    -0.06
    tx
    -0.06
    捕
    -0.06
    POSITIVE LOGITS
    ίκη
    0.07
    unction
    0.07
     اق
    0.06
    .It
    0.06
    /sh
    0.06
     χρό
    0.06
     mau
    0.06
     '*.
    0.06
     teslim
    0.06
    [data
    0.06
    Act Density 0.008%

    No Known Activations