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 activates on BEM‐style CSS class names using the double‐underscore “__” separator.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    洲
    -0.07
    epsilon
    -0.06
     violence
    -0.06
     Crus
    -0.06
    opak
    -0.06
     Guild
    -0.06
     Phillips
    -0.06
     collapse
    -0.06
     symbol
    -0.06
    ceed
    -0.06
    POSITIVE LOGITS
     Πα
    0.07
    (){
↵
↵
    0.07
    	se
    0.07
    .Des
    0.06
     Ej
    0.06
    各种
    0.06
    >);↵
    0.06
    目前
    0.06
    requete
    0.06
    elementType
    0.06
    Act Density 0.198%

    No Known Activations