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

    Neuronpedia

    INDEX
    Explanations

    beginning of string anchor

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    м
    3.64
    م
    3.21
    时候
    2.94
    ने
    2.89
    le
    2.80
    પ
    2.75
    ال
    2.68
     ongoing
    2.60
    DataObject
    2.54
    특별시
    2.52
    POSITIVE LOGITS
    '^
    4.64
    .^
    4.06
    "^
    3.74
    _^
    3.59
    ^^
    3.16
    ^*
    3.04
     跳转
    3.04
    ти
    3.00
    ⺀
    2.93
    <^
    2.87
    Act Density 0.060%

    No Known Activations