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

    Neuronpedia

    INDEX
    Explanations

    test measuring

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     পাকিস্তানী
    0.40
    effectiveness
    0.38
    ‐
    0.38
     Subhanahu
    0.37
    ానని
    0.37
    resol
    0.36
    fty
    0.36
    ಹು
    0.36
     efficiencies
    0.36
     irreparable
    0.36
    POSITIVE LOGITS
    定义
    0.48
     provided
    0.43
    版权
    0.43
    右手
    0.40
     provides
    0.40
     counsel
    0.40
    剧情
    0.38
    计算
    0.37
    云
    0.37
    做
    0.37
    Act Density 0.000%

    No Known Activations