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

    Neuronpedia

    INDEX
    Explanations

    If you are struggling

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     nombres
    0.66
     corrects
    0.63
     plus
    0.63
     rules
    0.63
     multiplication
    0.62
     lesser
    0.61
     waven
    0.60
     männer
    0.59
     beerCount
    0.59
     textura
    0.59
    POSITIVE LOGITS
    opportunities
    0.67
     amacıyla
    0.67
    Opportunities
    0.66
    возможно
    0.66
     Возможно
    0.65
    if
    0.65
    there
    0.63
    ou
    0.63
     hiện
    0.62
    As
    0.61
    Act Density 0.008%

    No Known Activations