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

    Neuronpedia

    INDEX
    Explanations

    Object, select

    np_max-act · gemini-2.0-flash

    This neuron fires whenever the text is inside a code snippet or shows programming commands (e.g. PowerShell/WMI invocation), effectively detecting code-block content.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    گذ
    -0.07
    imité
    -0.06
     KeyError
    -0.06
    "],["
    -0.06
    	mp
    -0.06
    ullah
    -0.06
     heartbeat
    -0.06
    <decimal
    -0.06
    -REAL
    -0.06
     iNdEx
    -0.06
    POSITIVE LOGITS
     mej
    0.07
     nghiệp
    0.06
     Lite
    0.06
     każ
    0.06
    conds
    0.06
     playback
    0.06
     overrun
    0.06
    pection
    0.06
     ön
    0.06
    Decor
    0.06
    Act Density 0.017%

    No Known Activations