说的是这个 s1模型的训练并非从零开始，而是基于阿里云通义千问（Qwen）模型进行监督微调。二级蒸馏

来源: Zhivago 于 2025-02-06 09:53:55 [旧帖] [给我悄悄话] 本文已被阅读：次

WENXUECITY.COM does not represent or guarantee the truthfulness, accuracy, or reliability of any of communications posted by other users.