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

    Neuronpedia

    INDEX
    Explanations

    describing attributes or entities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    5
    0.61
     conducta
    0.60
    8
    0.58
     análise
    0.57
    7
    0.57
    ност
    0.54
    ون
    0.52
     viser
    0.52
    வ்வேறு
    0.52
    و
    0.52
    POSITIVE LOGITS
     on
    0.64
     to
    0.61
     at
    0.55
     around
    0.54
     from
    0.53
     fiery
    0.52
     favored
    0.52
     rockets
    0.52
     McDonalds
    0.52
     flamboyant
    0.51
    Act Density 0.593%

    No Known Activations