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

    Neuronpedia

    INDEX
    Explanations

    high-risk, high-reward

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    sons
    0.51
    l
    0.49
    бат
    0.46
    x
    0.46
    lui
    0.46
    rv
    0.45
    time
    0.45
    p
    0.44
    fps
    0.44
    h
    0.44
    POSITIVE LOGITS
     consumes
    0.54
     consume
    0.50
     파일
    0.46
     Honolulu
    0.44
     consommation
    0.44
     गोनी
    0.43
     سٹی
    0.43
     personalize
    0.42
    损
    0.42
     해야
    0.42
    Act Density 0.004%

    No Known Activations