jianzhnie/LLamaTuner

Issues

• 该项目与qlora的差别

  #100 opened 6 months ago by lixiaoxiangzhi
  0

• 参数设置

  #102 opened 10 months ago by TheLolita
  0

• 不同样式的样本对应什么样的情形，如何根据自己的需求选择样本的样式

  #101 opened a year ago by lixiaoxiangzhi
  0

• 这么好的项目怎么没有issues，我先赞一个

  #3 opened a year ago by nieallen
  0

• QLORA微调alpaca_data.json报错 'padding_value' (position 3) must be float, not NoneType

  #96 opened a year ago by kanxueli
  2

• 总是这个错误怎么解决

  #93 opened a year ago by liumaohui
  1

• llama-2-13b的模型用单卡跑lora就会报错

  #95 opened a year ago by Enoch202
  1

• 批量推理时结果异常

  #98 opened a year ago by ZayIsAllYouNeed
  0

• 微调后的Llama-2-7b，在模型加载时出错

  #97 opened a year ago by kanxueli
  1

• faile on 3090

  #4 opened a year ago by SeekPoint
  6

• [问题]有关训练可视化

  #92 opened a year ago by RickMeow
  2

• About llama-2-70B fine-tuning

  #91 opened a year ago by RickMeow
  2

• zero3保存的模型无法加载

  #90 opened a year ago by charliedream1
  2

• 如何使用自己的数据集

  #89 opened a year ago by clannadcl
  1

• Tokenizer class BaiChuanTokenizer does not exist or is not currently imported.

  #11 opened a year ago by corlin
  6

• 下载了百川7b模型后，直接在gradio_webserver.py里推理，生成内容乱码问题

  #78 opened a year ago by FDwangchao
  3

• llama2-13B和llama2-70b微调所需要的显卡配置

  #87 opened a year ago by batindfa
  1

• 训练成功了，但是没有合并的脚本，请问如何合并？

  #7 opened a year ago by apachemycat
  4

• ValueError: Undefined dataset tatsu-lab/alpaca

  #81 opened a year ago by SeekPoint
  3

• 请问支持Baichuan 13B吗？

  #56 opened a year ago by mynewstart
  4

• 单机多卡并行训练报错

  #69 opened a year ago by wgzhendong
  3

• Baichuan7B使用lora微调后测试时总会再次输出query

  #73 opened a year ago by GaoXinJian-USTC
  1

• 多卡似乎不能将每张卡跑满，请问如何才能让每张卡的计算负载跑满呢

  #66 opened a year ago by RayneSun
  13

• Would you support RWKV?

  #57 opened a year ago by xiaol
  1

• 中文文档里没写对baichuan-13B的支持，但英文写了

  #62 opened a year ago by RayneSun
  1

• 微调训练失败

  #63 opened a year ago by QJShan
  4

• 多卡加速支持evalution吗

  #61 opened a year ago by greenriver777
  1

• data_utils.py 是不是有问题？

  #59 opened a year ago by wgzhendong
  1

• 32g内存+3060ti6G显存可以finetune 7B的模型吗？

  #54 opened a year ago by TodayWei
  1

• baichuan-7B: AttributeError: 'CastOutputToFloat' object has no attribute 'weight'

  #39 opened a year ago by franciszhang92
  1

• RuntimeError: probability tensor contains either `inf`, `nan` or element < 0

  #6 opened a year ago by weifan-zhao
  1

• 百川7B 模型微调结果

  #30 opened a year ago by jianzhnie
  4

• 数据预处理标签问题

  #24 opened a year ago by blackxer
  3

• How to adapt QLoRA to other base models?

  #9 opened a year ago by RanchiZhao
  4

• 4bit loaded error

  #2 opened a year ago by yuxuan2015
  3

• 合并的Bug修复，等会就出来验证结果了

  #14 opened a year ago by apachemycat
  0

• 合并代码还有一个错误，中间少了一个变量

  #12 opened a year ago by apachemycat
  1

• 8bit和4bit训练效果对比有吗

  #8 opened a year ago by kongjiellx
  1

• utils/apply_lora.py 有个小Bug

  #10 opened a year ago by apachemycat
  1

• RuntimeError: probability tensor contains either `inf`, `nan` or element < 0

  #5 opened a year ago by weifan-zhao
  4

• What's the real LICENSING

  #1 opened a year ago by jiacheo
  1