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

    Neuronpedia

    INDEX
    Explanations

    above percentage

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     দেড়
    0.56
    Half
    0.54
     ¼
    0.54
    十几
    0.53
    几十
    0.50
     HALF
    0.50
     dozens
    0.48
    tfrac
    0.48
     معظم
    0.47
     Half
    0.47
    POSITIVE LOGITS
     continuance
    0.39
     recurrence
    0.37
     семи
    0.37
     sev
    0.36
     cumulative
    0.35
     apod
    0.35
     itib
    0.35
    )}{\
    0.35
     complacency
    0.35
     >
    0.35
    Act Density 0.045%

    No Known Activations