Symbol bar c_tok

The passage around this formula

…across many targets when paid once as a supernet [ 10 ] , on top of whatever strategy trains the architecture it discovers. Sparse mixture-of-experts is the odd one out in this accounting: its saving shows up only in $\bar c_{\text{tok}}$ , the cost per served token, and it does essentially nothing to relax the constraint $M_{\text{resident}}$ $\le$ $B_{\text{mem}}$ that the other three strategies are built specifically to satisfy [ 13 , 12 ] .

Read this part in the article →