Struct Sampler 

pub struct Sampler { /* private fields */ }

A small, seedable xorshift64* generator. Not cryptographically secure – sampling doesn’t need that – but reproducible given a seed, which greedy argmax already was for free.

Implementations

impl Sampler

pub fn new(seed: u64) -> Self

pub fn sample( &mut self, logits: &[f32], params: &SamplingParams, history: &[usize], ) -> usize

Samples one token id from logits, given params and the already-generated history (for repetition penalty). Falls back to plain greedy argmax when params.temperature <= 0.0.

A length-1 logits vector is treated as a precomputed greedy token id (logits[0] as usize) — used by the Metal dense-stack path that returns GPU argmax instead of downloading the full vocab.

pub fn sample_with_mask( &mut self, logits: &[f32], params: &SamplingParams, history: &[usize], mask: Option<LogitMask<'_>>, ) -> usize

Like Self::sample, but optionally zeroes disallowed logits via mask before argmax / nucleus sampling (used for JSON-object mode).

pub fn uniform(&mut self) -> f32

A uniform draw in [0.0, 1.0).

Exposed because speculative decoding’s accept test is a coin flip against p_target(x) / p_draft(x) rather than a draw from a distribution, and it must come off the same seeded stream as every other draw in the run or a “reproducible given a seed” generation stops being reproducible.

pub fn sample_from(&mut self, probs: &[f32]) -> usize

Draws one index from an already-normalised distribution.

Split out of Self::sample_with_mask so speculative decoding can sample from a distribution it had to compute anyway (the rejection rule needs p_target itself, not just a draw from it) and still go through exactly the same draw as ordinary sampling. Two separate copies of this loop would be two chances to be subtly non-lossless.