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

    Neuronpedia

    INDEX
    Explanations

    percentage changes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ఉంటాయి
    0.64
    朊
    0.59
    长期
    0.58
     существу
    0.58
    循环
    0.57
     ఉంటుంది
    0.55
     нередко
    0.54
    悝
    0.54
    也不会
    0.54
    用来
    0.54
    POSITIVE LOGITS
     decrease
    0.80
     recorded
    0.80
     overall
    0.77
     hubo
    0.77
     YoY
    0.77
    decrease
    0.76
     increase
    0.76
     total
    0.75
     decreased
    0.74
     récord
    0.73
    Act Density 0.037%

    No Known Activations