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

    Neuronpedia

    INDEX
    Explanations

    Pennsylvania and Michigan universities

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     efficiencies
    -0.25
    æ©IJ
    -0.25
     deceive
    -0.25
    ä½³
    -0.25
    å¸Ŀçİĭ
    -0.24
    (runtime
    -0.24
    éŃį
    -0.24
    åݿ级
    -0.24
     Olympia
    -0.24
     crap
    -0.23
    POSITIVE LOGITS
    imately
    0.30
    stile
    0.28
    omanip
    0.28
    çıŃ
    0.27
    usi
    0.27
    漫
    0.27
    æ³Ľ
    0.27
    indo
    0.26
    MET
    0.25
    ä½ĵ
    0.25
    Act Density 0.007%

    No Known Activations