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

gated_group_rmsnorm_kernel

def gated_group_rmsnorm_kernel[dtype: DType, gate_dtype: DType, OutLayout: TensorLayout, YLayout: TensorLayout, GateLayout: TensorLayout, WeightLayout: TensorLayout, OutEngine: TensorEngine, YEngine: TensorEngine, GateEngine: TensorEngine, WeightEngine: TensorEngine](output: TileTensor[dtype, OutLayout, MutAnyOrigin, Engine=OutEngine], y: TileTensor[dtype, YLayout, ImmutAnyOrigin, Engine=YEngine], gate: TileTensor[gate_dtype, GateLayout, ImmutAnyOrigin, Engine=GateEngine], weight: TileTensor[.float32, WeightLayout, ImmutAnyOrigin, Engine=WeightEngine], n_rows_dev: Int32, num_groups_dev: Int32, group_size_dev: Int32, eps: Float32)