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

    Neuronpedia

    INDEX
    Explanations

    large quantities/long times

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     railway
    -0.07
    规定
    -0.07
     Strikes
    -0.07
     sounds
    -0.07
    _frames
    -0.07
     Exam
    -0.07
     alien
    -0.07
    Sound
    -0.07
    ophone
    -0.07
     Pennsylvania
    -0.07
    POSITIVE LOGITS
    伸出
    0.07
    𝓰
    0.07
    �
    0.07
    さら
    0.07
    hr
    0.06
     joked
    0.06
     RTAL
    0.06
    坬
    0.06
    𝙥
    0.06
    小时候
    0.06
    Act Density 0.022%

    No Known Activations