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

    Neuronpedia

    INDEX
    Explanations

    contrast and conjunctions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Four
    0.45
    خط
    0.44
     www
    0.44
     Pl
    0.42
     Which
    0.41
     This
    0.41
     key
    0.41
     ull
    0.40
     Five
    0.40
    Four
    0.39
    POSITIVE LOGITS
    이지만
    0.63
     çünkü
    0.61
     क्योंकि
    0.59
     karena
    0.58
     когда
    0.58
     แต่
    0.58
     porque
    0.57
     있지만
    0.57
     BECAUSE
    0.57
     પરંતુ
    0.57
    Act Density 0.002%

    No Known Activations