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

    Neuronpedia

    INDEX
    Explanations

    quoted text prefixes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ل
    3.50
    で
    3.27
    ти
    3.08
    야
    2.92
    ার
    2.86
    에서
    2.84
    ov
    2.67
    ज
    2.50
    어
    2.47
    ტ
    2.41
    POSITIVE LOGITS
    将
    2.11
    "{
    2.02
    外的
    2.02
     وعلى
    1.98
     এছাড়া
    1.87
    ""
    1.73
    User
    1.73
    sächlich
    1.73
    𝟘
    1.72
    "",
    1.66
    Act Density 0.220%

    No Known Activations