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

    Neuronpedia

    INDEX
    Explanations

    code and URLs

    np_max-act · gemini-2.0-flash

    This neuron detects the opening square bracket token (“[”) that marks the start of the answer or explanation placeholder in the prescribed output format.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     stools
    -0.07
    Procedure
    -0.06
    -edge
    -0.06
    _STRUCTURE
    -0.06
    Specifications
    -0.06
    Arn
    -0.06
    Terms
    -0.06
    γορ
    -0.06
    Young
    -0.06
    Ef
    -0.06
    POSITIVE LOGITS
    _candidates
    0.07
     rozh
    0.07
     customize
    0.07
    uggling
    0.07
     tham
    0.06
     nuevos
    0.06
     improvement
    0.06
    长
    0.06
     confidence
    0.06
    _agent
    0.06
    Act Density 0.003%

    No Known Activations