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

    Neuronpedia

    INDEX
    Explanations

    ，

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    看看
    -0.11
     yummy
    -0.11
     pretending
    -0.10
     lovely
    -0.10
     żeby
    -0.10
     messed
    -0.09
     pretend
    -0.09
     figuring
    -0.09
    Someone
    -0.09
     quelqu
    -0.09
    POSITIVE LOGITS
    ，实现
    0.09
     temporal
    0.08
     disparate
    0.08
    ?s
    0.08
     sämtliche
    0.08
     meticulously
    0.08
     exceeding
    0.08
     seamlessly
    0.08
     Temporal
    0.07
    699
    0.07
    Act Density 0.126%

    No Known Activations