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

    Neuronpedia

    INDEX
    Explanations

    use

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    -grid
    -0.07
     Gand
    -0.07
    多年前
    -0.07
     Fors
    -0.07
    猛地
    -0.07
    連結
    -0.07
    dd
    -0.07
    研发投入
    -0.07
     Coff
    -0.07
    决战
    -0.07
    POSITIVE LOGITS
    퓬
    0.08
    (Media
    0.07
     destroy
    0.07
    ,
    0.07
    (stack
    0.07
    .opts
    0.07
    鬓
    0.07
    	mock
    0.07
    (cluster
    0.07
    -mask
    0.07
    Act Density 0.013%

    No Known Activations