4 ms·
llama 3.x is already sharing the last layer with the embedding layer, it just uses the transpose in the last layer operation.
by danielmarkbruce 2y ago
llama 3.x is already sharing the last layer with the embedding layer, it just uses the transpose in the last layer operation.