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

    Neuronpedia

    INDEX
    Explanations

    large amounts and scale

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     kleines
    0.91
    smaller
    0.81
     неболь
    0.81
     kleinere
    0.80
     smaller
    0.80
     kleine
    0.79
    ව
    0.78
     kleiner
    0.76
     hubungan
    0.76
     suasana
    0.75
    POSITIVE LOGITS
     amounts
    1.99
     amount
    1.82
    amounts
    1.67
     quantities
    1.55
     scale
    1.55
    amount
    1.52
     Amounts
    1.46
     cantidades
    1.45
     quantidade
    1.42
    scale
    1.41
    Act Density 0.623%

    No Known Activations