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

    Neuronpedia

    INDEX
    Explanations

    Scaling AI Infrastructure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    deficient
    0.48
    ó
    0.46
    یند
    0.44
    ினா
    0.44
    返
    0.43
    というと
    0.43
    ían
    0.42
     conting
    0.42
     दुर्ग
    0.42
    required
    0.41
    POSITIVE LOGITS
     Walter
    0.54
     叶
    0.53
     双
    0.52
     Editar
    0.49
    ాప
    0.48
     verdens
    0.48
    Walter
    0.46
     ziyaret
    0.46
    버스
    0.45
     během
    0.45
    Act Density 0.002%

    No Known Activations