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

    Neuronpedia

    INDEX
    Explanations

    draining residual water/power

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Abdel
    -0.08
     donor
    -0.08
     variety
    -0.08
    选择
    -0.08
     Variety
    -0.07
     Taj
    -0.07
     bikin
    -0.07
     juxtap
    -0.07
     Morgan
    -0.07
     Salam
    -0.07
    POSITIVE LOGITS
    .flush
    0.11
    .Flush
    0.10
    _flush
    0.10
    Flush
    0.10
     flushed
    0.10
    flush
    0.10
     Flush
    0.10
     flush
    0.10
     ഒഴ
    0.10
     finalizar
    0.09
    Act Density 0.010%

    No Known Activations