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

    Neuronpedia

    INDEX
    Explanations

    clean

    np_max-act · gemini-2.0-flash

    This neuron activates on tokens related to cleaning or disinfecting actions (e.g., “net‐toyage,” “limpieza,” “desinfección,” “clean”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    าชน
    -0.07
     exhibition
    -0.06
    ult
    -0.06
    셜
    -0.06
    494
    -0.06
    pull
    -0.06
    xDB
    -0.06
    (StringUtils
    -0.06
    ULT
    -0.06
    expr
    -0.06
    POSITIVE LOGITS
     forCell
    0.07
    <SpriteRenderer
    0.07
     Morav
    0.07
    ')[
    0.07
    @Repository
    0.06
    /Data
    0.06
     createState
    0.06
    :"",↵
    0.06
    ")[
    0.06
     Armed
    0.06
    Act Density 0.022%

    No Known Activations