Struct RmsNormOp 

pub struct RmsNormOp {
    pub tokens: usize,
    pub dim: usize,
    pub eps: f32,
}

One concrete rms_norm invocation. Inputs:

• x: tokens × dim activation
• w: dim weight (per-channel scale)
• eps: the usual RMSNorm epsilon

Output: tokens × dim, same dtype as input on every backend’s compute dtype (typically fp16 on Metal/CUDA, fp32 on CPU).

Fields

tokens: usize

dim: usize

eps: f32

Trait Implementations

impl OpUnderTest for RmsNormOp

fn name(&self) -> &str

Display name (used in test failure messages).

fn run_cpu(&self, seed: u64) -> Output

Run on CPU (reference). Always available.