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
ReduceRMSNormFusedResidualAdd
struct ReduceRMSNormFusedResidualAdd
Registers the mo.composite.rms_norm_fused_residual_add graph op with the graph compiler.
Implemented traits
Methods
execute
static def execute[dtype: DType, rank: Int, target: StringSpan[ImmStaticOrigin], epsilon1: Float32, epsilon2: Float32, weight_offset1: Float32, weight_offset2: Float32, multiply_before_cast: Bool = True](output: ManagedTensorSlice[IOSpec[_, _].Output, static_spec=output.static_spec], residual_output: ManagedTensorSlice[IOSpec[_, _].Output, static_spec=residual_output.static_spec], input: ManagedTensorSlice[IOSpec[_, _].FusedInput, static_spec=input.static_spec], residual_input: ManagedTensorSlice[IOSpec[_, _].FusedInput, static_spec=residual_input.static_spec], gamma1: ManagedTensorSlice[IOSpec[_, _].Input, static_spec=gamma1.static_spec], gamma2: ManagedTensorSlice[IOSpec[_, _].Input, static_spec=gamma2.static_spec], ctx: DeviceContextBase)
Executes the mo.composite.rms_norm_fused_residual_add graph op.
Parameters:
- dtype (
DType): Element type of the input and output tensors. - rank (
Int): Tensor rank of the input and output tensors. - target (
StringSpan[ImmStaticOrigin]): Compilation target string. - epsilon1 (
Float32): Stability constant for the first normalization. - epsilon2 (
Float32): Stability constant for the second normalization. - weight_offset1 (
Float32): Scalar offset for the first weight. - weight_offset2 (
Float32): Scalar offset for the second weight. - multiply_before_cast (
Bool): See the graph op signature.
Args:
- output (
ManagedTensorSlice[IOSpec[_, _].Output, static_spec=output.static_spec]): Output tensor receiving the result. - residual_output (
ManagedTensorSlice[IOSpec[_, _].Output, static_spec=residual_output.static_spec]): See the graph op signature. - input (
ManagedTensorSlice[IOSpec[_, _].FusedInput, static_spec=input.static_spec]): Input tensor to reduce. - residual_input (
ManagedTensorSlice[IOSpec[_, _].FusedInput, static_spec=residual_input.static_spec]): Residual tensor added to the normalized input. - gamma1 (
ManagedTensorSlice[IOSpec[_, _].Input, static_spec=gamma1.static_spec]): Scale weights for the first normalization. - gamma2 (
ManagedTensorSlice[IOSpec[_, _].Input, static_spec=gamma2.static_spec]): Scale weights for the second normalization. - ctx (
DeviceContextBase): Device context used to enqueue the kernel.
Raises:
Error: If the operation parameters are invalid.