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

    Neuronpedia

    INDEX
    Explanations

    city names, place names, GT, 6

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ği
    3.20
    ு
    3.10
    𝙞
    3.09
    leqslant
    2.99
    𝙮
    2.91
    ឧ
    2.85
    ference
    2.78
     dass
    2.76
    وں
    2.75
    ंट
    2.69
    POSITIVE LOGITS
    로
    3.58
    3.07
    萏
    2.77
    ારીખ
    2.65
    ार्किक
    2.64
    に
    2.63
    𝚘
    2.61
     persuaded
    2.57
     же
    2.55
    tama
    2.54
    Act Density 0.111%

    No Known Activations