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

    Neuronpedia

    INDEX
    Explanations

    time units

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Layers
    0.95
     Years
    0.91
    Years
    0.86
    layers
    0.85
     Layers
    0.85
     layers
    0.82
    Stages
    0.79
    Steps
    0.78
     Rows
    0.76
    多年的
    0.76
    POSITIVE LOGITS
     month
    2.49
     week
    2.34
    month
    1.99
     season
    1.86
    week
    1.82
     year
    1.76
     Month
    1.72
     quarter
    1.56
    Month
    1.55
     mês
    1.50
    Act Density 0.210%

    No Known Activations