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

    Neuronpedia

    INDEX
    Explanations

    text cleaning

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    нулся
    -0.08
    -0.08
    Crystal
    -0.08
     ruhig
    -0.08
     überhaupt
    -0.08
     Kindes
    -0.08
     biling
    -0.08
     Crystal
    -0.07
     affirm
    -0.07
     உண
    -0.07
    POSITIVE LOGITS
    掉
    0.17
     unnecessary
    0.12
     лиш
    0.11
     unwanted
    0.11
     excessive
    0.10
    _duplicates
    0.10
     undes
    0.10
     pesky
    0.09
    Duplicates
    0.09
     undue
    0.09
    Act Density 0.017%
    No Known Activations