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

    Neuronpedia

    INDEX
    Explanations

    Duplication/replacement

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Differences
    -0.09
     differences
    -0.08
    んでいる
    -0.07
    見える
    -0.07
    ewhat
    -0.07
     illusions
    -0.07
    -this
    -0.07
     Pig
    -0.07
    ↵            ↵
    -0.07
    ↵            
↵
    -0.07
    POSITIVE LOGITS
    ナン
    0.08
    Opening
    0.08
    ERAL
    0.07
    WARNING
    0.07
    ন
    0.07
    쾌
    0.07
    網
    0.07
    ༄
    0.07
     tearDown
    0.07
    	select
    0.07
    Act Density 0.046%

    No Known Activations