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

    Neuronpedia

    INDEX
    Explanations

    puckering or stamps

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Burgers
    0.69
     중심
    0.64
     Rewrite
    0.63
     Shrimp
    0.63
     Routing
    0.63
     Brothers
    0.63
     XPath
    0.63
     Broccoli
    0.60
     Förderung
    0.60
     Selenium
    0.60
    POSITIVE LOGITS
    U
    0.66
    for
    0.60
    ici
    0.55
    orn
    0.55
    awk
    0.55
    sm
    0.55
    used
    0.54
    uf
    0.54
    attempts
    0.54
    samples
    0.54
    Act Density 0.001%

    No Known Activations