Top-p (Nucleus) Sampling

Top-p (nucleus) sampling selects the smallest set of tokens whose cumulative probability exceeds threshold p, then samples from them. Unlike top-k (fixed count), top-p adapts: it picks fewer tokens when the model is confident and more when uncertain.

Steps:

1. Apply softmax to logits
2. Sort probabilities descending
3. Compute cumulative sum
4. Keep tokens up to and including the one that crosses threshold p
5. Renormalize and return in original token order

Signature: def top_p_sampling(logits, p)

• logits: (vocab_size,) — unnormalized scores
• p: float — cumulative probability threshold (0 < p ≤ 1)
• Returns: (vocab_size,) — renormalized probability distribution