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

    Neuronpedia

    INDEX
    Explanations

    python code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    事実
    0.73
    的任务
    0.72
    任務
    0.72
    operations
    0.71
     عملی
    0.71
     kompet
    0.71
    長的
    0.71
    教學
    0.71
    業
    0.70
    थित
    0.69
    POSITIVE LOGITS
     Patio
    0.80
     patio
    0.80
     tavern
    0.74
     parmesan
    0.73
     torso
    0.72
     alpine
    0.71
     rear
    0.71
     rectangular
    0.71
     loam
    0.70
     superficies
    0.70
    Act Density 0.332%

    No Known Activations