All great ideas are simple
That’s why we got the reuse principle at work.
Run a reuse computation
If you have a 48‐dimensional embedding, you have 48 “features” describing each token’s characteristics.
这里可以举例说明，什么是 feature
GPT-3 (175 billion-parameter): C=12,288
Embedding 是把 token 的 位置信息 以及 某个语义特征结合起来
1. mean (μ) and variance (Var)
2. The epsilon term (ε = 1×10-5) is there to prevent division by zero.
3. 特征修正
When instruction ignores these cognitive realities, school delivers chronic frustration instead of satisfying insight. Align lessons with how the mind learns—manageable problems, concrete examples, cumulative knowledge, and practiced fluency—and curiosity has room to flourish.
1. To gain minimum competence
2. To gain proficiency
3. It helps the mental process become automatic.
4. It makes memory long-lasting.
5. It increases transfer of learning to new situations.
memorization of core knowledge is a necessary foundation for deeper skills
如果你的日语单词好了，很多事情可能就自发的产生了
也就是说记忆单词（多邻国）是一个非常好的活化能刺激器
其实就是两步走，每两步都很快乐，但是一起进行就不快乐
记忆词汇，疼，但有动作游戏化的感觉
思考推理，疼，但如果词汇比较熟悉，令人快乐，有解谜游戏化