-
|
如题， |
Beta Was this translation helpful? Give feedback.
Answered by
ymcui
Oct 26, 2023
Replies: 1 comment 3 replies
-
|
训练什么参数取决于训练脚本，和“预训练”、“精调”这些词无关。 |
Beta Was this translation helpful? Give feedback.
3 replies
-
|
感谢回答。我明白提供的代码是LoRA训练，请问发布的Chinese模型(llama2 & alpaca2)也都是LoRA训练后的结果吗？还是全量预训练的结果？ |
Beta Was this translation helpful? Give feedback.
-
|
是。论文里有相关资料。FAQ里也有解答（LINK）。 |
Beta Was this translation helpful? Give feedback.
-
|
好的谢谢 |
Beta Was this translation helpful? Give feedback.
Answer selected by
RedZh
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
训练什么参数取决于训练脚本，和“预训练”、“精调”这些词无关。
这里说的“预训练”指的是无标注数据训练，“精调”指的是“指令精调”，即使用指令数据进行有监督的训练。
我们提供的代码主要以lora训练为主。如果你想全量预训练就自己改改代码吧。