Closed Yaziwel closed 6 months ago
Have you incorporated a bidirectional attention mechanism into RWKV6, or are you adhering to the unidirectional attention mechanism described in the original paper?
Thanks for your attention! VRWKV6 is also bidirectional attention like VRWKV.
Have you incorporated a bidirectional attention mechanism into RWKV6, or are you adhering to the unidirectional attention mechanism described in the original paper?