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

    Neuronpedia

    INDEX
    Explanations

    introduces contrast

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     toutefois
    0.59
    겠지만
    0.57
     però
    0.56
     allerdings
    0.56
     ancak
    0.52
    但
    0.52
     cependant
    0.51
     কিন্ত
    0.51
     tuttavia
    0.50
    However
    0.49
    POSITIVE LOGITS
     of
    0.54
    like
    0.48
    ternut
    0.46
    jeća
    0.46
     that
    0.44
    ht
    0.43
    ts
    0.42
    tiny
    0.42
     Ꭰ
    0.42
     /,
    0.42
    Act Density 0.032%

    No Known Activations