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

    Neuronpedia

    INDEX
    Explanations

    How comfortable are you

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    is
    0.76
    at
    0.69
    it
    0.68
    ow
    0.67
    if
    0.63
    ak
    0.62
    int
    0.62
    im
    0.61
    us
    0.61
    ad
    0.60
    POSITIVE LOGITS
    ProductsStore
    0.52
     pathos
    0.51
     vassals
    0.51
     nascent
    0.49
     school
    0.47
     sap
    0.47
     تھا۔
    0.47
     ሳይ
    0.47
     membahas
    0.47
     partisans
    0.47
    Act Density 0.147%

    No Known Activations