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

    Neuronpedia

    INDEX
    Explanations

    degradation/progression over time

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    _gain
    -0.07
    פיד
    -0.07
     intern
    -0.07
     immigr
    -0.07
    Nos
    -0.07
    总产值
    -0.07
    �
    -0.07
     NEC
    -0.06
    ᗜ
    -0.06
    면
    -0.06
    POSITIVE LOGITS
    .story
    0.07
    minster
    0.07
    subplot
    0.07
    (AT
    0.07
     rests
    0.07
    创作
    0.07
    undle
    0.07
    [row
    0.07
    _authenticated
    0.07
     convicted
    0.07
    Act Density 0.235%

    No Known Activations