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

    Neuronpedia

    INDEX
    Explanations

    how something is

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    เพราะ
    0.77
    если
    0.73
    因为
    0.71
     eftersom
    0.71
    เนื่อง
    0.71
    隨著
    0.70
     apesar
    0.70
     क्योंकि
    0.69
     因为
    0.69
     protože
    0.68
    POSITIVE LOGITS
     directly
    0.66
     (
    0.64
     squarely
    0.59
     indirectly
    0.59
     को
    0.58
     unequivocally
    0.55
     successively
    0.55
     solely
    0.55
     effectively
    0.54
     uniquely
    0.53
    Act Density 0.333%

    No Known Activations