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

    Neuronpedia

    INDEX
    Explanations

    Pinyin and Chinese place names

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Osaka
    0.71
    缦
    0.70
    にな
    0.66
     Tokyo
    0.66
     Burlington
    0.66
     Kyoto
    0.65
     epoch
    0.65
    மாகக்
    0.64
     Git
    0.64
     Lizard
    0.63
    POSITIVE LOGITS
    qi
    1.00
    zi
    1.00
    ying
    0.99
    mei
    0.96
    wei
    0.96
    xi
    0.94
     wei
    0.92
    fei
    0.90
    zhi
    0.90
    fa
    0.90
    Act Density 0.044%

    No Known Activations