DSpark Qwen3 版本的模型默认忽略 `partial_rotary_factor`, 导致检查点的配置与训练权重不匹配(导致 vLLM 服务出现问题)

def apply_rotary_pos_emb(q, k, cos, sin, unsqueeze_dim=1):
    ...
    q_embed = (q * cos[..., -q_len:, :]) + (rotate_half(q) * sin[..., -q_len:, :])
    k_embed = (k * cos) + (rotate_half(k) * sin)
- L239 — `self.rotary_emb = Qwen3RotaryEmbedding(config)`. With transformers 5.13.1 and a config declaring `partial_rotary_factor: 0.25`, `head_dim: 128`, this still yields **128-dim** cos/sin (`inv_freq` len 64), so the declared factor never takes effect at training time either way.