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

    Neuronpedia

    INDEX
    Explanations

    Japan, Japanese, ume, Memory Lane

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    鶘
    0.83
    鎊
    0.79
    颶
    0.77
    猞
    0.74
    瘓
    0.71
    騾
    0.71
    劊
    0.69
    妫
    0.69
    虢
    0.68
    🌯
    0.68
    POSITIVE LOGITS
     Japanese
    2.75
     Japan
    2.69
     일본
    2.53
    Japanese
    2.44
     япон
    2.42
    Japan
    2.39
     japan
    2.34
     Tokyo
    2.27
     Jepang
    2.23
     japanese
    2.22
    Act Density 0.234%

    No Known Activations