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

    Neuronpedia

    INDEX
    Explanations

    coverage

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    Mt
    -0.08
    In
    -0.08
    ki
    -0.07
     boş
    -0.07
    utch
    -0.07
    ואב
    -0.07
     salt
    -0.07
     sour
    -0.07
     convention
    -0.07
    _BLOCKS
    -0.06
    POSITIVE LOGITS
     Coverage
    0.08
    僚
    0.07
    %%%%%%%%%%%%%%%%
    0.07
    flight
    0.07
    提升了
    0.07
    可能是
    0.07
    פור
    0.07
     piracy
    0.07
    确保
    0.07
    _upgrade
    0.07
    Act Density 0.006%

    No Known Activations