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

    Neuronpedia

    INDEX
    Explanations

    introducing examples

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     반드시
    0.84
     myös
    0.78
     आभार
    0.77
     обов
    0.77
     także
    0.77
     শীঘ্রই
    0.76
     mutlaka
    0.74
     serta
    0.73
    的选择
    0.71
    纷纷
    0.70
    POSITIVE LOGITS
     imagine
    3.82
    Imagine
    3.62
     Imagine
    3.55
    imagine
    3.31
     Suppose
    2.81
     suppose
    2.69
     think
    2.69
    Suppose
    2.68
     imagines
    2.66
    想象
    2.56
    Act Density 0.678%

    No Known Activations