转移字符的情况下，\=。
楼主说不行，我还专门测试了下。
a = '{"code":"A00185","data":"\u5bf9\u4e0d\u8d77\uff0c\u60a8\u77ed\u65f6\u95f4\u53d1\u8868\u535a\u6587\u8fc7\u591a\uff0c\u8bf7\u591a\u4f11\u606f\uff0c\u6ce8\u610f\u8eab\u4f53\uff01\u611f\u8c22\u60a8\u5bf9\u65b0\u6d6a\u535a\u5ba2\u7684\u652f\u6301\u548c\u5173\u6ce8\uff01"}'
b = a.replace('\','hehehehe')
print(b);
结果是
{"code":"A00185","data":"heheheheu5bf9heheheheu4e0dheheheheu8d77heheheheuff0cheheheheu60a8heheheheu77edheheheheu65f6heh
heheu95f4heheheheu53d1heheheheu8868heheheheu535aheheheheu6587heheheheu8fc7heheheheu591aheheheheuff0cheheheheu8bf7hehehe
eu591aheheheheu4f11heheheheu606fheheheheuff0cheheheheu6ce8heheheheu610fheheheheu8eabheheheheu4f53heheheheuff01heheheheu
11fheheheheu8c22heheheheu60a8heheheheu5bf9heheheheu65b0heheheheu6d6aheheheheu535aheheheheu5ba2heheheheu7684heheheheu652
heheheheu6301heheheheu548cheheheheu5173heheheheu6ce8heheheheuff01"}
看到楼主新帖的代码片段，再更一下，楼主没见代码提示都变红了？
a.replace('\','')
这句本来就是错误的写法，因为'转义将一个右引号转义了之后，第二个字符串就不完整了，当然会报错。
另外，按照题主的要求，本来就不需要替换，因为这个字符串本来就是一个，在定义字符串的时候，写两个\，才等于真正的一个。
可以用以下代码验证：
a = '{"code":"A00185","data":"\u5bf9\u4e0d\u8d77\uff0c\u60a8\u77ed\u65f6\u95f4\u53d1\u8868\u535a\u6587\u8fc7\u591a\uff0c\u8bf7\u591a\u4f11\u606f\uff0c\u6ce8\u610f\u8eab\u4f53\uff01\u611f\u8c22\u60a8\u5bf9\u65b0\u6d6a\u535a\u5ba2\u7684\u652f\u6301\u548c\u5173\u6ce8\uff01"}'
print(a);
输出结果
{"code":"A00185","data":"u5bf9u4e0du8d77uff0cu60a8u77edu65f6u95f4u53d1u8868u535au6587u8fc7u591auff0cu8bf
7u591au4f11u606fuff0cu6ce8u610fu8eabu4f53uff01u611fu8c22u60a8u5bf9u65b0u6d6au535au5ba2u7684u652fu630
1u548cu5173u6ce8uff01"}
几番交涉，终于知道了题主的目的，就是想解析出json中的中文而已。。。
很简单
import json
a = '{"code":"A00185","data":"\u5bf9\u4e0d\u8d77\uff0c\u60a8\u77ed\u65f6\u95f4\u53d1\u8868\u535a\u6587\u8fc7\u591a\uff0c\u8bf7\u591a\u4f11\u606f\uff0c\u6ce8\u610f\u8eab\u4f53\uff01\u611f\u8c22\u60a8\u5bf9\u65b0\u6d6a\u535a\u5ba2\u7684\u652f\u6301\u548c\u5173\u6ce8\uff01"}'
js = json.loads(a)
print (js)
输出
{'data': '对不起，您短时间发表博文过多，请多休息，注意身体！感谢您对新浪博客的支持和关注！', 'code': 'A00185'}
