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

    Neuronpedia

    INDEX
    Explanations

    programming and applications

    np_max-act · gemini-2.0-flash

    This neuron detects the labels and keywords used to invoke or describe tool‐use actions (e.g. tool names like “Search” or “Calculator” and action keys like “action” or “action_input”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Canterbury
    -0.07
     Ur
    -0.06
     Sung
    -0.06
    <const
    -0.06
     Early
    -0.06
     Arcade
    -0.06
    _raise
    -0.06
     nghỉ
    -0.06
    """
↵
    -0.06
     concurrently
    -0.06
    POSITIVE LOGITS
     bouquet
    0.08
     гост
    0.07
    ots
    0.07
    repository
    0.06
    odd
    0.06
    -a
    0.06
    _kernel
    0.06
    -il
    0.06
    účast
    0.06
     dood
    0.06
    Act Density 0.023%

    No Known Activations