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

    Neuronpedia

    INDEX
    Explanations

    struct

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    Makes
    -0.09
     Notíc
    -0.07
    汶川
    -0.07
    	BufferedReader
    -0.07
    ield
    -0.07
    swith
    -0.06
     <?=
    -0.06
    掬
    -0.06
     Anc
    -0.06
     sàn
    -0.06
    POSITIVE LOGITS
    ਊ
    0.07
     vấn
    0.07
     loophole
    0.07
    ptive
    0.07
    دام
    0.07
     recharge
    0.06
     awakened
    0.06
    훽
    0.06
    ップ
    0.06
     picker
    0.06
    Act Density 0.002%

    No Known Activations