Function rms_norm_strided 

pub fn rms_norm_strided(
    gpu: &dyn GpuBackend,
    kernel: KernelHandle,
    input: DevicePtr,
    weight: &DenseWeight,
    output: DevicePtr,
    rows_per_group: u32,
    num_groups: u32,
    hidden_size: u32,
    eps: f32,
    row_stride: u32,
    stream: u64,
) -> Result<()>

2026-09-25: RMS norm of num_groups groups of rows_per_group rows in one launch. Groups start row_stride elements apart, and rows inside a group are packed at hidden_size. The multi-sequence q/k head norms use it because each sequence’s heads sit inside that sequence’s own interleaved block.

rms_norm_strided in kernels/gb10/common/rms_norm.cu has the per-row body of rms_norm there; only the row’s base address differs.