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

    Neuronpedia

    INDEX
    Explanations

    why, what, where

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Deviation
    0.77
    ा
    0.67
    ه
    0.64
     भंग
    0.64
     uneven
    0.61
     wishes
    0.60
     waxaa
    0.58
    uut
    0.58
     whim
    0.57
     desviación
    0.57
    POSITIVE LOGITS
    ídas
    0.70
     wikip
    0.70
    Mbps
    0.67
    ído
    0.66
     proviene
    0.64
     gesagt
    0.63
    琲
    0.62
    ेद
    0.62
    ето
    0.62
     μεγά
    0.61
    Act Density 0.349%

    No Known Activations