For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /get-started.md).
Mojo struct
Struct_moe_single_group_router_eplb
struct Struct_moe_single_group_router_eplb
Registers the mo.moe.single.group.router.eplb graph op with the graph compiler.
Implemented traits
Methods
execute
static def execute[scores_type: DType, bias_type: DType, //, n_routed_experts: Int, n_experts_per_tok: Int, norm_weights: Bool, num_log: Int, max_replicas: Int, hash_decorrelate: Bool, target: StringSlice[ImmStaticOrigin]](expert_indices: ManagedTensorSlice[IOSpec[_, _].Output, static_spec=expert_indices.static_spec], expert_indices_log: ManagedTensorSlice[IOSpec[_, _].Output, static_spec=expert_indices_log.static_spec], expert_weights: ManagedTensorSlice[IOSpec[_, _].Output, static_spec=expert_weights.static_spec], expert_scores: ManagedTensorSlice[IOSpec[_, _].FusedInput, static_spec=expert_scores.static_spec], expert_bias: ManagedTensorSlice[IOSpec[_, _].Input, static_spec=expert_bias.static_spec], logcnt: ManagedTensorSlice[IOSpec[_, _].Input, static_spec=logcnt.static_spec], log2phy: ManagedTensorSlice[IOSpec[_, _].Input, static_spec=log2phy.static_spec], layer_idx: ManagedTensorSlice[IOSpec[_, _].Input, static_spec=layer_idx.static_spec], routed_scaling_factor: Float32, context: DeviceContext)