Score each group using its strongest two adjusted affinities

Read the equation

The 256 experts are arranged as eight groups of thirty-two. Within each token and group, take the two largest bias-adjusted scores and add them. The resulting eight group scores decide which groups are eligible for the final expert selection.

Terms & definitions

How it works

Read the definition from its right-hand expression to the quantity on the left. Its surrounding source records the intended input, state, or parameter context. Where the source leaves an operation unspecified, this page preserves that gap.

Assumptions & limits

Source recovery establishes that the expression appears in this record. Its mathematical validity, implementation behavior, and interpretation must be assessed separately. See the linked source and research notes for the checks available for this version.

Original context

3.  scores_for_choice = scores + e_score_correction_bias            -> (T, 256)
      The F32 (256,) bias, broadcast over the token dimension.

  4.  group_scores = scores_for_choice.view(T, 8, 32).topk(2, -1)[0].sum(-1)
                                                                      -> (T, 8)
      Each of the 8 groups is scored by THE SUM OF ITS TOP TWO members.

  5.  group_idx = topk(group_scores, k=4)                             -> (T, 4)

group_scores = scores_for_choice.view(T, 8, 32).topk(2, -1)[0].sum(-1)

Related builds