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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    This neuron fires on numeric literals with decimal points (floating-point constants) in code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    	engine
    -0.08
     files
    -0.07
    rypt
    -0.06
     forward
    -0.06
     propaganda
    -0.06
     neckline
    -0.06
    pliance
    -0.06
    	graph
    -0.06
     Soy
    -0.06
    	cv
    -0.06
    POSITIVE LOGITS
     тобі
    0.07
    Winner
    0.07
     Integral
    0.07
    -year
    0.06
     तत
    0.06
    _valor
    0.06
    BASH
    0.06
     setTime
    0.06
     aylık
    0.06
    umhur
    0.06
    Act Density 0.111%

    No Known Activations