This is the first Chinese chat model specifically fine-tuned for Chinese through ORPO based on the Meta-Llama-3-8B-Instruct model. - Shenzhi-Wang/Llama3-Chinese-Chat
This is the first Chinese chat model specifically fine-tuned for Chinese through ORPO based on the Meta-Llama-3-8B-Instruct model. - Llama3-Chinese-Chat/README.md at main · Shenzhi-Wang/Llama3-Chinese-Chat
Compared to the original Meta-Llama-3.1-8B-Instruct model, our llama3.1-8B-Chinese-Chat model significantly reduces the issues of "Chinese questions with English answers" and the mixing of Chinese and English in responses. The training dataset contains >100K preference pairs, and it exhibits signi...
Shenzhi WangShenzhi-Wang Block or Report PinnedLoading Llama3-Chinese-ChatLlama3-Chinese-ChatPublic This is the first Chinese chat model specifically fine-tuned for Chinese through ORPO based on the Meta-Llama-3-8B-Instruct model. 30717
--output_dir /nfsdata/gpu007/dingfei/commondata/huggingface/Meta-Llama-3-8B-Instruct-saves/shenzhi-wang/Llama3-8B-Chinese-Chat-v1-test1 --per_device_train_batch_size 2 --per_device_eval_batch_size 2 --gradient_accumulation_steps 4 --lr_scheduler_type cosine --log_level info ...
你好作者,我在使用djl环境下运行了Llama3-8B-Chinese-Chat-q4_0-v2_1.gguf模型,推理的结果部分单词会发生乱码, 我通过debug代码发现一个UTF8的字节数组拆成两个字节数组,正常'奋'的UFT8字节编码为[-27, -91, -117]。
在我的测试中,您所用的prompt应该就不会出现夹带英文的情况了。 👍 1 Owner Shenzhi-Wang commented May 6, 2024 可以尝试一下我们的v2.1的版本,这个问题减少了很多。 https://huggingface.co/shenzhi-wang/Llama3-8B-Chinese-Chat/commit/4788ab8512511daa7b80f75c85ceb703661a4a4cSign...
This is the first Chinese chat model specifically fine-tuned for Chinese through ORPO based on the Meta-Llama-3-8B-Instruct model. - 有gradio那种网页版的部署方式吗? · Issue #4 · Shenzhi-Wang/Llama3-Chinese-Chat
This is the first Chinese chat model specifically fine-tuned for Chinese through ORPO based on the Meta-Llama-3-8B-Instruct model. - context_length是微调成2k的了吗 · Issue #3 · Shenzhi-Wang/Llama3-Chinese-Chat