FeiyangChen

Results 2 issues of FeiyangChen

We add support for - different hidden dimension between qk and v. - not equal num_heads_k and num_heads_v, such as (num_heads_q, num_heads_k, num_heads_v) = (32, 4, 16). For different hidden...

How do I implement a kernel with the arguments a list of tensors? When implementing the grouped gemm kernel, I want the weight input to be a list of tensors....

enhancement