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

    Neuronpedia

    INDEX
    Explanations

    the

    np_max-act · gemini-2.0-flash

    This neuron detects numerical values formatted as decimal numbers (i.e., floating-point figures).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Breath
    -0.08
     Rift
    -0.07
     rowspan
    -0.07
    .quantity
    -0.07
     ********
    -0.07
     compliment
    -0.07
    multiline
    -0.06
     caracter
    -0.06
     Fakat
    -0.06
     Pepper
    -0.06
    POSITIVE LOGITS
    !');↵
    0.07
    eated
    0.07
    +");↵
    0.06
    ?id
    0.06
    %@
    0.06
     Equipment
    0.06
    Outlet
    0.06
    	rv
    0.06
     ()↵
    0.06
    视频
    0.06
    Act Density 0.014%

    No Known Activations