git-cloner / llama2-lora-fine-tuning

llama2 finetuning with deepspeed and lora

https://gitclone.com/aiit/chat/

git-cloner/llama2-lora-fine-tuning Issues

ImportError: cannot import name 'import_path' from '_pytest.doctest'
Updated a month ago1
deepspeed这里是使用zero-1/2/3哪个模式切片的？
Updated 3 months ago
为什么generate里面使用base model的tokenizer
Updated 4 months ago
ValueError: Attention mask should be of size (4, 1, 240, 480), but is torch.Size([4, 1, 240, 240])
Updated 5 months ago3
decoder输出长度是有限制吗？
Updated 7 months ago5
stream_output 是不是没用到呀
Updated 7 months ago1
MultiGPU+Deepspeed+4bitQlora
Updated 9 months ago1
===================================BUG REPORT===================================
Closed 10 months ago
可以支持多机多卡吗，如何配置支持deepspeed的stage2，stage3
Closed 10 months ago
pip install git+https://github.com/huggingface/transformers -i https://pypi.mirrors.ustc.edu.cn/simple”这步报错了
Closed 10 months ago1
可以支持多机多卡吗
Closed 10 months ago2
validation_files
Updated 10 months ago1
llama2-13B和llama2-70b所需要的显卡配置
Updated 10 months ago1
请教关于微调
Closed 10 months ago5
请问扩充中文词表的作用是什么呀
Closed 10 months ago2