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

    Neuronpedia

    INDEX
    Explanations

    twice

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    清洁
    -0.07
     evening
    -0.07
    汛
    -0.07
    沼
    -0.07
    恶性
    -0.07
     nostalg
    -0.06
    뺄
    -0.06
    羰
    -0.06
     Barbar
    -0.06
     seminar
    -0.06
    POSITIVE LOGITS
    жи
    0.07
    +W
    0.07
    什么东西
    0.07
    	↵↵
    0.07
     "`
    0.07
     twice
    0.06
    🤣
    0.06
    素材
    0.06
    ().↵
    0.06
     OTHERWISE
    0.06
    Act Density 0.004%

    No Known Activations