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

    Neuronpedia

    INDEX
    Explanations

    rewarding endeavors and experiences

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ’
    1.09
    is
    0.94
    大
    0.91
    1
    0.90
    án
    0.90
    '
    0.84
    um
    0.83
    puted
    0.82
    2
    0.82
    5
    0.82
    POSITIVE LOGITS
     rewarding
    1.13
    ו
    1.09
     worthwhile
    1.04
     rewards
    1.02
     rewarded
    0.90
    lardan
    0.90
     enjoyable
    0.89
    ться
    0.88
     arduous
    0.88
     immensely
    0.87
    Act Density 0.013%

    No Known Activations