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

    Neuronpedia

    INDEX
    Explanations

    temporal sequences and conditions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    )
    0.23
    0.23
     กับ
    0.21
    ),
    0.21
     위해
    0.20
    '
    0.20
    _
    0.20
     six
    0.19
    ’
    0.19
     如果
    0.19
    POSITIVE LOGITS
     embarking
    0.23
    писы
    0.21
     finalizing
    0.19
    жи
    0.18
    embark
    0.18
    Ketika
    0.18
     accessing
    0.17
    ки
    0.17
    ња
    0.17
    人们
    0.17
    Act Density 0.486%

    No Known Activations