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

    Neuronpedia

    INDEX
    Explanations

    reflections and broken things

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     действия
    -0.07
     действий
    -0.07
    (channel
    -0.07
     পালন
    -0.07
     เช่น
    -0.07
    elser
    -0.07
    ုပ်
    -0.07
     caliber
    -0.07
    (selection
    -0.07
     <<<
    -0.07
    POSITIVE LOGITS
     fragmented
    0.11
     damaged
    0.11
     vieja
    0.10
     hurried
    0.10
    旧
    0.10
     broken
    0.10
    broken
    0.10
    Broken
    0.10
    Recover
    0.10
     battered
    0.10
    Act Density 0.097%

    No Known Activations