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

    Neuronpedia

    INDEX
    Explanations

    work, bubbling, specific combinations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    rs
    0.44
    \
    0.41
     waxed
    0.40
    )
    0.40
    $
    0.39
    ‬
    0.39
    shaw
    0.39
    ssel
    0.39
    delta
    0.38
    bd
    0.38
    POSITIVE LOGITS
     concernant
    0.54
    ന്റ്
    0.50
     ಮೆ
    0.50
     работаю
    0.50
     थोरो
    0.50
     도시
    0.48
     conductors
    0.48
    ໃຫ້
    0.48
     prevede
    0.48
    meli
    0.48
    Act Density 0.004%

    No Known Activations