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

    Neuronpedia

    INDEX
    Explanations

    Increasing/decreasing intensity

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    ům
    -0.07
    	operator
    -0.06
     sued
    -0.06
    与
    -0.06
    ег
    -0.06
     advis
    -0.06
     pear
    -0.06
     ル
    -0.06
    endforeach
    -0.06
    Perl
    -0.06
    POSITIVE LOGITS
     veri
    0.07
     Tracks
    0.07
    words
    0.06
     относят
    0.06
     zákona
    0.06
    Scientists
    0.06
    !!!↵
    0.06
    交通
    0.06
     signatures
    0.06
    нее
    0.06
    Act Density 0.067%

    No Known Activations