Mojo function

batched_matmul_dynamic_scaled_fp8_naive

batched_matmul_dynamic_scaled_fp8_naive[c_type: DType, a_type: DType, b_type: DType, a_scales_type: DType, b_scales_type: DType, //, *, scales_granularity_mnk: IndexList[3], transpose_b: Bool = False](c_: LayoutTensor[c_type, c_.layout, c_.origin, address_space=c_.address_space, element_layout=c_.element_layout, layout_int_type=c_.layout_int_type, linear_idx_type=c_.linear_idx_type, masked=c_.masked, alignment=c_.alignment], a_: LayoutTensor[a_type, a_.layout, a_.origin, address_space=a_.address_space, element_layout=a_.element_layout, layout_int_type=a_.layout_int_type, linear_idx_type=a_.linear_idx_type, masked=a_.masked, alignment=a_.alignment], b_: LayoutTensor[b_type, b_.layout, b_.origin, address_space=b_.address_space, element_layout=b_.element_layout, layout_int_type=b_.layout_int_type, linear_idx_type=b_.linear_idx_type, masked=b_.masked, alignment=b_.alignment], a_scales_: LayoutTensor[a_scales_type, a_scales_.layout, a_scales_.origin, address_space=a_scales_.address_space, element_layout=a_scales_.element_layout, layout_int_type=a_scales_.layout_int_type, linear_idx_type=a_scales_.linear_idx_type, masked=a_scales_.masked, alignment=a_scales_.alignment], b_scales_: LayoutTensor[b_scales_type, b_scales_.layout, b_scales_.origin, address_space=b_scales_.address_space, element_layout=b_scales_.element_layout, layout_int_type=b_scales_.layout_int_type, linear_idx_type=b_scales_.linear_idx_type, masked=b_scales_.masked, alignment=b_scales_.alignment], ctx: DeviceContext)