Symbol m_2

The passage around this formula

…a smaller one — a reported example decomposes an “Einstein” feature into components resembling “scientist,” “Germany,” and “famous person.” Writing $d_i^{(m_2)}$ for one feature direction recovered by a dictionary of size $m_2$ and $d_j^{(m_1)}$ for the directions recovered by a smaller dictionary of size $m_1$ , their finding is that

Read this part in the article →