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 function
composite_rms_norm_fused_residual_add_shape
def composite_rms_norm_fused_residual_add_shape(input: T, residual_input: T, gamma1: T, gamma2: T) -> IndexList[T.rank]
Computes the output shape for the mo.composite.rms_norm_fused_residual_add graph op.
Args:
- βinput (
T): Primary input tensor whose shape the output mirrors. - βresidual_input (
T): Residual tensor added to the normalizedinput. - βgamma1 (
T): Per-column scale weights applied to the first RMS normalization. - βgamma2 (
T): Per-column scale weights applied to the second RMS normalization.
Returns:
IndexList[T.rank]: The output shape, which matches the input shape.