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

    Neuronpedia

    INDEX
    Explanations

    exactly

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    .socket
    -0.07
    .val
    -0.07
     Musk
    -0.07
    pell
    -0.07
     choke
    -0.07
    天才
    -0.07
     Polo
    -0.07
    
    -0.06
     Evalu
    -0.06
    (exit
    -0.06
    POSITIVE LOGITS
     tj
    0.07
    ||↵
    0.07
    ordable
    0.07
     Swamp
    0.07
     marrying
    0.07
     laying
    0.06
    AVAILABLE
    0.06
    married
    0.06
    。
↵
    0.06
    为主的
    0.06
    Act Density 0.008%

    No Known Activations