Update to moe w/ hybrid attention

Author: Rohan-Bierneni

docs/reference/models/supported_models_and_architectures.md

@@ -44,7 +44,7 @@ MaxText is an open-source, high-performance LLM framework written in Python/JAX.
 
 ### Qwen3
 
-- **Variants**: Dense (0.6B–32B); MoE (30B-A3B, 235B-A22B, 480B Coder), MoE w/ Hybrid (Next-80B-a3b)
+- **Variants**: Dense (0.6B–32B); MoE (30B-A3B, 235B-A22B, 480B Coder), MoE w/ Hybrid Attention (Next-80B-a3b)
 - **Notes**: **QK-Norm**, GQA, SwiGLU, RMSNorm, RoPE, GatedDeltaNet.
 
 ### GPT-OSS