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

    Neuronpedia

    INDEX
    Explanations

    washed and added

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    权力
    0.60
     repurposed
    0.58
     المُ
    0.57
    机器学习
    0.57
    Sadly
    0.56
    掌控
    0.56
    总结
    0.55
    架构
    0.54
    运维
    0.54
     हिंसा
    0.54
    POSITIVE LOGITS
     prepd
    0.90
     thereof
    0.80
     wherein
    0.72
     owing
    0.71
    leqq
    0.70
     thereto
    0.68
     thereon
    0.63
     constituting
    0.62
     therefrom
    0.60
     hereinafter
    0.60
    Act Density 0.005%

    No Known Activations