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

    Neuronpedia

    INDEX
    Explanations

    old understanding

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ма
    0.87
     assured
    0.80
     전
    0.75
    ла
    0.75
     chased
    0.73
     го
    0.73
     топо
    0.73
     前
    0.69
     unavailable
    0.68
    м
    0.68
    POSITIVE LOGITS
    whitespace
    0.83
    Determ
    0.76
    ającym
    0.74
    alış
    0.73
    üyor
    0.73
    boundingRect
    0.73
    Conclusions
    0.72
    jectory
    0.71
     Determining
    0.71
    cimientos
    0.70
    Act Density 0.000%

    No Known Activations