Xenova HF Staff

sync 6fdf6301e2bb

5ec57d5 verified about 15 hours ago

5.87 kB

library_name: kernels
license: apache-2.0
tags:
  - kernel
  - webgpu
  - wgsl

com.microsoft.SkipSimplifiedLayerNormalization

com.microsoft · ONNX Runtime contrib operator · contrib since_version 1

Description

Adds input and skip (plus optional bias), then applies RMS normalization scaled by gamma. The optional second output exposes the pre-normalization sum. The schema's training-only mean and inverse-standard-deviation outputs are not implemented.

See the ONNX Runtime SkipSimplifiedLayerNormalization contrib-operator spec for the reference semantics.

Inputs

Name	Upstream name	Logical dtype	Rank	Shape	Description	Presence
`inputT`	`input`	`T`	—	—	Input tensor of shape `(token_count, hidden_size)` or `(batch, sequence, hidden_size)`, normalized over the last axis.	required
`skipT`	`skip`	`T`	—	—	Residual tensor of the same shape as `input`, added before normalization.	required
`gammaT`	`gamma`	`T`	`1`	—	1-D scale tensor with shape `(hidden_size)` applied after normalization.	required
`biasT`	`bias`	`T`	`1`	—	Optional 1-D bias tensor with shape `(hidden_size)` added to the `input + skip` sum.	optional

Outputs

Name	Upstream name	Logical dtype	Rank	Shape	Description	Presence
`outputT`	`output`	`T`	same as `inputT`	same as `inputT`	Normalized output tensor with the same shape as `input`.	required
`meanT`	`mean`	`U`	same as `inputT`	derived	Per-row mean; zero for simplified RMS normalization. Shape matches the input with its final axis replaced by one.	optional
`invStdT`	`inv_std_var`	`U`	same as `inputT`	derived	Per-row inverse standard deviation, or inverse RMS for simplified normalization. Shape matches the input with its final axis replaced by one.	optional
`residualT`	`input_skip_bias_sum`	`T`	same as `inputT`	same as `inputT`	Sum of `input`, `skip`, and optional `bias` before normalization, with the same shape as `input`.	optional

Attributes

Default values (overridable per request):

Attribute	Default	Description
`epsilon`	`9.999999960041972e-13`	Non-negative epsilon added to the mean square before taking the square root.

Type constraints

Variable	Allowed dtypes
`T`	`float32`, `float16`
`U`	`float32`

Implementation variants

One implementation is selected per call from the device capabilities, the request shapes and the dtypes; these notes say what each one covers.

stats_mean_plain — Row normalization returning mean statistics with plain optional inputs and outputs.
stats_mean_residual — Row normalization returning mean statistics with residual optional inputs and outputs.
stats_mean_bias — Row normalization returning mean statistics with bias optional inputs and outputs.
stats_mean_bias_residual — Row normalization returning mean statistics with bias_residual optional inputs and outputs.
stats_inv_plain — Row normalization returning inv statistics with plain optional inputs and outputs.
stats_inv_residual — Row normalization returning inv statistics with residual optional inputs and outputs.
stats_inv_bias — Row normalization returning inv statistics with bias optional inputs and outputs.
stats_inv_bias_residual — Row normalization returning inv statistics with bias_residual optional inputs and outputs.
stats_both_plain — Row normalization returning both statistics with plain optional inputs and outputs.
stats_both_residual — Row normalization returning both statistics with residual optional inputs and outputs.
stats_both_bias — Row normalization returning both statistics with bias optional inputs and outputs.
stats_both_bias_residual — Row normalization returning both statistics with bias_residual optional inputs and outputs.

Device requirements

Some implementation variants require shader-f16. These are route-specific capabilities, not package-wide requirements; availability also depends on the request shape and dtype.

Files

metadata.json — kernel metadata (id, digests, per-variant templates, provenance)
manifest.json — the op contract (source of truth)
test.json — correctness cases
bench.json — benchmark cases
norm-skip-row-vec4.wgsl.jinja
norm-skip-row.wgsl.jinja

Use with `@huggingface/kernels`

npm install --save-exact @huggingface/kernels@0.0.1-preview.3

Required output shapes and logical data types are inferred from the supplied inputs and attributes; result tensors are allocated automatically.

The version: 1 option selects the published kernel contract; it is independent of any operator opset, contrib since_version, or model version. It follows the v1 branch as fixes land. To pin exact artifact bytes, pass a 40-character commit revision instead of version.

Replace each *Data placeholder with a typed array containing the corresponding input data.

import { getKernel } from "@huggingface/kernels";

const kernel = await getKernel("webgpu-kernels/com.microsoft.SkipSimplifiedLayerNormalization", { version: 1 });
const { outputT } = await kernel({
  inputT: { data: inputTData, shape: [2, 4] },
  skipT: { data: skipTData, shape: [2, 4] },
  gammaT: { data: gammaTData, shape: [4] },
});