Update link in deterministic blog (#207)

* Update link in deterministic

* upd

Author: Fridge003

blog/2025-09-22-sglang-deterministic.md

@@ -160,7 +160,7 @@ Attention kernel is an important part of determinism. For different attention ba
 
 
 ### Reproducible Non-Greedy Sampling
-To extend determinism beyond greedy decoding, we introduce a new sampling function: [multinomial_with_seed](https://github.com/sgl-project/sglang/blob/fb1e8acd2954b6267c73a199427976d89887ff0e/python/sglang/srt/layers/sampler.py#L263).
+To extend determinism beyond greedy decoding, we introduce a new sampling function: [multinomial_with_seed](https://github.com/sgl-project/sglang/blob/e2ac7888b8cb1fd6c33a7ec58d27a5f5b5b24e0c/python/sglang/srt/layers/sampler.py#L268-L299).
 
 Instead of relying on `torch.multinomial`, which is inherently nondeterministic under batching, this operator perturbs logits with Gumbel noise generated from a **seeded hash function**. As a result, the same `(inputs, seed)` pair always yields the same sample, even when temperature > 0.