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

    Neuronpedia

    INDEX
    Explanations

    np. followed by function

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    larda
    2.80
    ية
    2.25
    lty
    2.17
     nghề
    2.14
    lardan
    2.13
    ค์
    1.98
     Đảng
    1.97
     וע
    1.94
    m
    1.88
    φό
    1.83
    POSITIVE LOGITS
    ك
    2.95
    说
    2.83
    ک
    2.73
    д
    2.17
    ль
    2.13
    ्ञ
    2.11
    خ
    2.11
    ES
    2.09
    ructions
    2.08
    はこの
    2.05
    Act Density 0.011%

    No Known Activations