git-cloner llama2-lora-fine-tuning issues - Githubissues

git-cloner / llama2-lora-fine-tuning

llama2 finetuning with deepspeed and lora

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

MIT License

155 stars 14 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

deepspeed这里是使用zero-1/2/3哪个模式切片的？

#15 rechawine opened 3 months ago
0
ImportError: cannot import name 'import_path' from '_pytest.doctest'

#14 boolmriver opened 3 months ago
1
为什么generate里面使用base model的tokenizer

#13 kunzeng-ch opened 5 months ago
0
ValueError: Attention mask should be of size (4, 1, 240, 480), but is torch.Size([4, 1, 240, 240])

#12 LiBinNLP opened 6 months ago
3
decoder输出长度是有限制吗？

#11 MarsMeng1994 opened 8 months ago
5
stream_output 是不是没用到呀

#10 MarsMeng1994 opened 8 months ago
1
MultiGPU+Deepspeed+4bitQlora

#9 yaoching0 opened 10 months ago
1
===================================BUG REPORT===================================

#8 backatonesJ closed 10 months ago
0
pip install git+https://github.com/huggingface/transformers -i https://pypi.mirrors.ustc.edu.cn/simple”这步报错了

#7 backatonesJ closed 10 months ago
1
可以支持多机多卡吗，如何配置支持deepspeed的stage2，stage3

#6 yangzhipeng1108 closed 10 months ago
0
可以支持多机多卡吗

#5 yangzhipeng1108 closed 10 months ago
2
validation_files

#4 wchengyu opened 10 months ago
1
llama2-13B和llama2-70b所需要的显卡配置

#3 batindfa opened 11 months ago
1
请教关于微调

#2 goog closed 11 months ago
5
请问扩充中文词表的作用是什么呀

#1 goog closed 11 months ago
2