For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /max/get-started.md).
Mojo struct
RMSNormFusedQuantizeDynamicScaledFP8
struct RMSNormFusedQuantizeDynamicScaledFP8
Implemented traitsβ
AnyType,
ImplicitlyDestructible
Methodsβ
executeβ
static def execute[input_dtype: DType, output_dtype: DType, scale_dtype: DType, rank: Int, target: StringSlice[StaticConstantOrigin]](output: ManagedTensorSlice[Output, static_spec=output.static_spec], scales: ManagedTensorSlice[Output, static_spec=scales.static_spec], input: ManagedTensorSlice[FusedInput, static_spec=input.static_spec], gamma: ManagedTensorSlice[Input, static_spec=gamma.static_spec], epsilon: Scalar[input_dtype], weight_offset: Scalar[input_dtype], scale_ub: Float32, ctx: DeviceContext)
shapeβ
static def shape[input_dtype: DType, rank: Int](input: ManagedTensorSlice[Input, static_spec=input.static_spec], gamma: ManagedTensorSlice[Input, static_spec=gamma.static_spec], epsilon: Scalar[input_dtype], weight_offset: Scalar[input_dtype], scale_ub: Float32) -> IndexList[rank]
Returns:
Was this page helpful?
Thank you! We'll create more content like this.
Thank you for helping us improve!