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

    Neuronpedia

    INDEX
    Explanations

    Delimiters and separators

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     XM
    -0.07
    erman
    -0.07
     arbeit
    -0.07
     Conse
    -0.07
     Instructions
    -0.07
    verbose
    -0.07
     $,
    -0.07
     opaque
    -0.07
    symbols
    -0.07
     Filename
    -0.07
    POSITIVE LOGITS
     одинаков
    0.17
     regardless
    0.14
     irrespective
    0.14
    Regardless
    0.13
     независимо
    0.13
     одина
    0.13
     dezelfde
    0.12
     একই
    0.12
     Regardless
    0.11
     동일
    0.11
    Act Density 0.100%

    No Known Activations