Pinned issues
For issues releated to Qwen1.5, please report to https://github.com/QwenLM/Qwen1.5
#1097 opened by jklj077
Closed0
Issues
- 1
请问可以使用高通的npu进行部署和推理吗?
#1240 opened by caramel678 - 1
- 2
File "finetune.py", line 412, in <module> train() File "finetune.py", line 384, in train model = get_peft_model(model, lora_config) File "/opt/conda/envs/qwen/lib/python3.8/site-packages/peft/mapping.py", line 123, in get_peft_model peft_config.base_model_name_or_path = model.__dict__.get("name_or_path", None) AttributeError: 'NoneType' object has no attribute '__dict__'[BUG] <title>
#1248 opened by sunyclj - 5
[BUG] <Qwen-14B-Chat 输入长文本时无输出结果>
#1232 opened by TianWuYuJiangHenShou - 2
微调完成后使用llama_factory的vllm和qwen官方的vllm部署方式启动返回的不一样
#1241 opened by lxb0425 - 4
本地部署后,运行很慢啊
#1245 opened by wangwm - 2
qwen 14b 不微调的情况下,问相同的问题,模型输出也不太一致,是为什么?温度已经设置成0了
#1249 opened by Wangqi12138 - 7
[BUG] <title>lora微调loss异常?
#1214 opened by estuday - 3
💡 [REQUEST] - <title> 关于lora 模型合并的几个问题
#1226 opened by wangyao123456a - 2
[BUG] <title> 如何用vllm部署qlora后的模型
#1197 opened by KevinFan0 - 1
Qwen/eval中的评测CEval和CMMLU,开大推理的batchsize评测指标会显著降低
#1223 opened by ntz2000 - 4
[BUG] lora微调后,合并成一个模型。这种方式如何加载且推理
#1220 opened by wangyao123456a - 2
pip install csrc/layer_norm 不成功
#1208 opened by niykx - 1
如何添加`LogitsProcessor`控制结果输出?
#1213 opened by ChamfersChen - 1
- 1
请问下 2.5什么时候开源呀?
#1246 opened by xiguadong - 2
[HELP] I wonder how the MMLU result is evaluated?
#1189 opened by YuMeng2v - 1
💡 [REQUEST] - <title>数据集构造方法请教
#1191 opened by skyantao - 4
- 1
- 1
💡 [REQUEST] - <使用ollama来调用qwen:14B时,怎么设置输出文本长度呢>
#1243 opened by chaoskklt - 2
tokenizer.decoder 抛出'utf-8' codec can't decode bytes in position 1-2: unexpected end of data异常
#1218 opened by gaokao123 - 1
请问哪里可以找到qwen用于vllm的jinja template?
#1236 opened by xudong2019 - 1
- 2
- 2
[BUG] <关于model.generate时发现的源码错误>
#1231 opened by malidong521 - 4
请问基于qwen-72b-chat,基于怎样的配置可以在一台4090上训练起来?
#1224 opened by taishan1994 - 1
[BUG] Qwen/Qwen-72B-Chat-Int8,不能多GPU并行计算
#1222 opened by gquanma - 2
[BUG] <title> wrong system prompt check?
#1209 opened by XpastaX - 2
以lora、bfloat16方式微调模型,模型微调后采用lora参数和基座模型进行推理,使用merge_and_unload()类前后推理结果不一致,为什么会出现这种情况呢
#1168 opened by shaojh1 - 1
how to convert qwen.tiktoken to tokenzier.model
#1204 opened by cloudyuyuyu - 3
[BUG] .CalledProcessError: Command '['/usr/bin/gcc', '/tmp/tmpecd6su1w/main.c'
#1199 opened by ff1Zzd - 4
[BUG] CUDA Error: invalid device function /tmp/pip-req-build-5rlg4jgm/ln_fwd_kernels.cuh 236
#1198 opened by taoqinghua - 1
Run Qwen /openai_api.py, Error :Input should be a valid string, body.messages[3].function_call,请问Qwen1.5不支持了么?
#1205 opened by goqw - 0
- 1
openai_api.py启动的时候添加了username/password, 然后调用的时候怎么传入username/password呢?
#1190 opened by zengqingfu1442 - 2
- 1
请教,qlora微调没效果
#1188 opened by huangyunxin - 1
May I ask when Lisa fine-tuning is supported?
#1187 opened by zhaoyinjiang9825 - 1
微调过程中的taskType参数
#1186 opened by 12915494174 - 1
[BUG] <title>cannot import name 'allow_in_graph' from partially initialized module 'torch._dynamo' (most likely due to a circular import) (/demo/miniconda3/envs/qwen/lib/python3.9/site-packages/torch/_dynamo/__init__.py)
#1173 opened by dshwei - 1
在使用多卡做Qwen-7B-Chat做微调过程中出现ValueError: Expected a string path to an existing deepspeed config, or a dictionary, or a base64 encoded string. Received: finetune/ds_config_zero3.json
#1171 opened by TuDaCheng - 1
7B模型推理时生成非有效的idx,应该怎么处理?
#1174 opened by AACengineer - 5
[BUG] 对qwen-7b模型微调后,输出句子断句不正常,直接从句子中间停止
#1178 opened by twwch - 1
Calculate language probabilities
#1180 opened by itlittlekou - 5
4张卡为什么没有并发推理
#1179 opened by linzm1007 - 2
请问计划什么时间支持让vllm-gptq运行Qwen-72B-Chat-INT8大模型?
#1176 opened by linklogis-copilot - 6
deepspeed 单机多卡训练报错
#1175 opened by GXKIM - 1
qwen-14B-chat-int8/4 vllm模式部署错误:no kernel image is available for execution on the device
#1169 opened by Lzzycola - 1
損失函數計算(評估)方式
#1167 opened by swilly0906