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

    Neuronpedia

    INDEX
    Explanations

    asking to know

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     unfairly
    0.80
     certainly
    0.77
     unnatural
    0.77
     lackluster
    0.76
     unreasonable
    0.74
     would
    0.72
     unreasonably
    0.71
     probablemente
    0.70
     cube
    0.68
     suble
    0.68
    POSITIVE LOGITS
     удалось
    0.77
     così
    0.75
    如此
    0.73
    这么
    0.73
    こう
    0.71
    這麼
    0.70
     így
    0.69
    matmul
    0.67
     soweit
    0.67
     udało
    0.67
    Act Density 0.157%

    No Known Activations