datawhalechina/self-llm

Issues

• deepseek lora

  #91 opened 4 days ago by xionghao132
  2

• 【非Issues！讨论帖】通过lora微调的qianwen和直接使用system来预设的区别貌似不是很大

  #100 opened a month ago by Roger-G
  3

• 请问有多模态LLM的部署/微调文档吗，未来有相关更新计划吗

  #121 opened 20 days ago by ybshaw
  1

• peft微调llama3 8b，从第10补开始loss一直都是0

  #124 opened 14 days ago by ykallan
  2

• 微调Qwen1.5-0.5b报错 PermissionError: [Errno 13] Permission denied: './output/Qwen1.5\checkpoint-100'

  #123 opened 17 days ago by ykallan
  5

• 使用 llama3 的 lora 微调报错：NotImplementedError: Cannot copy out of meta tensor; no data!

  #115 opened a month ago by chenmonster
  3

• llama3 api报错

  #122 opened 19 days ago by samprasgit
  1

• 【XVERSE-7B-chat WebDemo 部署】报错 torch.cuda.OutOfMemoryError: CUDA out of memory.

  #113 opened a month ago by hgsw
  2

• deepseek-v2部署请求

  #120 opened 23 days ago by airsxue
  2

• 04-Qwen-7B-Chat Lora 微调时报错

  #119 opened 25 days ago by FisherAndHunter
  1

• 在纯 CPU 上可以运行吗？比如苹果电脑没有 cuda?

  #117 opened 25 days ago by aa03146
  1

• chatglm3,lora微调报错

  #116 opened 25 days ago by Jsonzhang20
  1

• 我在微调LLAMA3的时候出现NotImplementedError: Cannot copy out of meta tensor; no data!

  #110 opened a month ago by Mr-Lnan
  8

• Qwen1.5-7B Lora微调报错：RuntimeError: element 0 of tensors does not require grad and does not have a grad_fn

  #112 opened a month ago by feifeifei-hue
  2

• llama3 API调用的问题

  #114 opened a month ago by 1148514800
  1

• LLaMA3-8B-Instruct+lora使用A800（80GB显存）微调长度8192

  #93 opened 2 months ago by 12915494174
  2

• Qwen1.5-7B-Chat vLLM 部署调用-速度测试 hf命令错误

  #101 opened a month ago by ericalduo
  1

• 微调出来会有不礼貌或攻击性的言语

  #108 opened a month ago by 1737686924
  3

• Qwen1.5-7B Lora微调报错

  #107 opened a month ago by JeckerWen
  1

• qwen1.5 量化版本的部署方案有么？

  #54 opened a month ago by chongqingwei
  3

• 好兄弟 帮大忙了

  #50 opened a month ago by whm233
  0

• Building wheel for flash-attn (setup.py) ... - 卡住了。

  #99 opened a month ago by zhangyc
  4

• peft训练完成，参考的是04-Qwen-7B-Chat Lora 微调.ipynb，但是重新载入模型时，提示peft版本问题，用的是model = AutoModelForCausalLM.from_pretrained("../output/Qwen/checkpoint-1300/", trust_remote_code=True).eval()，提示的错误时ValueError: The version of PEFT you are using is not compatible, please use a version that is greater than 0.5.0

  #95 opened a month ago by Roger-G
  2

• InternLM2 缺少包

  #98 opened a month ago by JinProton
  2

• 报错asyncio.run() cannot be called from a running event loop，辛苦大佬们看看

  #97 opened 2 months ago by chaoskklt
  6

• 想问下在这个项目下的lora微调和Chatglm3官方微调的demo的数据格式怎么不一样呀

  #96 opened 2 months ago by McRays
  1

• 请问LLAMA3，里面是按1.2.3.4的顺序来分别执行吗？

  #94 opened 2 months ago by YadiHe
  6

• qwen-vl

  #84 opened 2 months ago by lyc728
  0

• 请问ChatGLM3微调中的数据集huanhuan的在哪获取？

  #81 opened 2 months ago by haory-95
  3

• 多卡报错，Qwen1.5-7B-Chat FastApi 部署调用

  #79 opened 2 months ago by linzhonghong
  3

• 请问chatglm模型Lora微调完成之后，如何加载新模型？

  #72 opened 2 months ago by waynetest2024
  8

• ChatGLM3-6B微调后成哑巴了（字面意思）

  #65 opened 3 months ago by Tangent-90C
  4

• 与FastChat的区别

  #78 opened 2 months ago by deeomnjobs
  0

• 建议按照顺序撰写README部分教程

  #75 opened 2 months ago by EuthL
  1

• chatglm3-6b fastapi调用

  #74 opened 2 months ago by caixiongjiang
  1

• Qwen-1.5-4B LLM推理bug

  #73 opened 2 months ago by caixiongjiang
  3

• 关于模型的api部署，能否推出高性能的异步版本、例如使用vllm、或者fastchat等工具

  #70 opened 2 months ago by lzh1998-jansen
  1

• deepseek官方readme，loss第二轮开始就是

  #63 opened 3 months ago by liuyongjie985
  7

• 我想我可不可提交给PR支持一下BlueLM我们的蓝心大模型

  #66 opened 3 months ago by acwwt
  1

• chatglm搭建知识库读取文件出错

  #53 opened 3 months ago by AHPUymhd
  3

• 帮大忙了

  #51 opened 3 months ago by sssggghhh007
  3

• 想问下部署chatglm出现了问题

  #55 opened 3 months ago by DJzing
  5

• 大佬这是什么情况？

  #60 opened 3 months ago by Jonnys-Li
  0

• 这不来个llama的微调教程

  #56 opened 3 months ago by Tangent-90C
  2

• Qwen1.5-7B推理部分，为什么我在modelscope上的GPU调试，fastapi报错。

  #58 opened 3 months ago by tking007
  1

• chatGLM3保存模型失败，报错TypeError: Object of type set is not JSON serializable

  #52 opened 3 months ago by BojackMa
  4

• 参考DeepSeek-7B-chat Lora 微调脚本微调deepseek-coder-7b-v1.5版本模型，生成的内容全是感叹号

  #48 opened 3 months ago by ivanzfb
  3

• chatGLM3微调过程中报错TypeError: Object of type set is not JSON serializable

  #47 opened 4 months ago by CharlieZZss
  3

• Failed to import transformers.models.qwen2

  #44 opened 4 months ago by lhtpluto
  8

• 代码笔误pd.read_json('../dataset/huanhuan.jsonl')，应该是pd.read_json('../dataset/huanhuan.json')

  #45 opened 4 months ago by Weihong-Liu
  2