Discussions

• 2 You must be logged in to vote

  🙏

  从transformers 自动下载模型，response, history = model.chat(tokenizer, "你好", history=[])报错too many values to unpack (expected 2)

  LiuXingEmail asked Jul 27, 2024 in Q&A · Unanswered

  

  

  0
• 1 You must be logged in to vote

  🙏

  关于工具参数的问题

  zykilled asked Jun 14, 2024 in Q&A · Unanswered

  

  

  0
• 1 You must be logged in to vote

  ☢️

  After update latest,The output is not accurate enough.

  haowaiwai started Jun 19, 2024 in Bad Case

  

  

  0
• 1 You must be logged in to vote

  🙏

  lora微调时，将batch_size调为64后，学习率应该怎么调，我的学习率10点多，下不去。并且训练途中在验证数据集测试时，模型输出为空，然后就会报错：

  Jasmine0224 asked Jun 25, 2024 in Q&A · Unanswered

  

  

  0
• 1 You must be logged in to vote

  🙏

  按官方文档，对ChatGLM3的LoRA微调之后，输出到output目录下的checkpoint是微调和基础模型合并了的模型吗？

  easychu asked Jun 26, 2024 in Q&A · Unanswered

  

  

  1
• 1 You must be logged in to vote

  🙏

  为什么不能从LangChain中加载ChatGLM3

  DerekAI asked Jun 22, 2024 in Q&A · Unanswered

  

  

  0
• 1 You must be logged in to vote

  🙏

  第一次尝试 github 中 lora_finetune.ipynb 微调示例，未执行成功，报以下错误，请问如何解决？

  thomasyyang asked Jun 24, 2024 in Q&A · Unanswered

  

  

  1
• 1 You must be logged in to vote

  🙏

  运行微调到2000条的时候突然报错

  F2023888 asked Jul 5, 2024 in Q&A · Unanswered

  

  

  0
• 1 You must be logged in to vote

  🙏

  在云服务部署之后，执行api_server.py报错No sentence-transformers model found with name BAAI/bge-m3. Creating a new one with mean pooling.

  Zackaxng asked Jul 11, 2024 in Q&A · Unanswered

  

  

  0
• 1 You must be logged in to vote

  💬

  Lora微调之后，如何进行rouge和bleu评估？

  Sunguy-me started Jul 23, 2024 in General

  

  

  0
• 1 You must be logged in to vote

  🙏

  在Jetson AGX Orin上进行4bit量化时报错

  FanZhang91 asked Jul 15, 2024 in Q&A · Unanswered

  

  

  2
• 1 You must be logged in to vote

  🙏

  FlashAttention only supports Ampere GPUs or newer.

  Lin-A1 asked Jul 16, 2024 in Q&A · Unanswered

  

  

  0
• 1 You must be logged in to vote

  🙏

  langchian 中 没有llm.bind_tools的方法。

  fengyutt asked Jul 16, 2024 in Q&A · Unanswered

  

  

  0
• 1 You must be logged in to vote

  🙏

  Local deployment error:ValueError: too many values to unpack (expected 2)

  SaltedCarl asked Jul 11, 2024 in Q&A · Unanswered

  

  

  1
• 1 You must be logged in to vote

  ☢️

  训练模型的时候，设置保存次数为200次，第200次和400次正常保存，然后训练到底500次的时候就会报错：AttributeError: 'NoneType' object has no attribute 'pop'

  casamchen started Jul 21, 2024 in Bad Case

  

  

  0