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

    Neuronpedia

    INDEX
    Explanations

    improvement and resolution

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    。
    2.11
    ،
    1.95
    يا
    1.90
    ，
    1.81
    стика
    1.75
    ：
    1.73
    目光
    1.65
    ές
    1.63
    ্ধু
    1.59
     ollut
    1.59
    POSITIVE LOGITS
    s
    3.11
    sion
    2.13
    the
    2.08
    ात
    1.97
    and
    1.90
     volna
    1.89
    sman
    1.78
    gend
    1.77
    ের
    1.73
    sant
    1.73
    Act Density 0.065%

    No Known Activations