issues
search
QwenLM
/
Qwen
The official repo of Qwen (通义千问) chat & pretrained large language model proposed by Alibaba Cloud.
Apache License 2.0
13.59k
stars
1.11k
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
[BUG] vLLM推理Qwen-14b-chat出现个别乱码和个别字母无法生成
#974
toufunao
closed
8 months ago
2
stop_words_ids
#973
zhaodongh
closed
5 months ago
2
[BUG] <title> 1.8B 模型 RAG 出现幻觉 + 回复 bug
#972
1100111GTH
closed
5 months ago
2
14Bint4量化后回答有乱码,无法使用
#971
GLY-123
closed
6 months ago
1
各位大佬 GPU功耗很低 但是GPU利用率满载 是什么情况[BUG] <title>
#970
oho-work
closed
8 months ago
4
72b全量微调300M语料时间80h正常不?
#969
lyc202001
closed
5 months ago
2
[BUG] | ERROR | stderr | vllm_worker.py: error: argument --model-path: expected one argument
#968
lhtpluto
closed
8 months ago
2
[BUG] 通过transformers对base model进行全量微调时loss归零
#967
Sniper970119
closed
8 months ago
5
[BUG] qwen1.8b微调后 调用模型tokenizer报错ModuleNotFoundError: No module named 'transformers_modules.qwen_1'
#966
tungsten106
closed
8 months ago
2
Fix bug of low_cpu_mem_usage in finetune.py.
#964
fyabc
closed
8 months ago
0
[BUG] <一机多卡V100lora微调卡住>
#963
yoohhwz
closed
6 months ago
9
[BUG] <title>openai_api.py 使用 stream=True 没有作用。
#962
rsghll
closed
8 months ago
2
[微调报错] <bash finetune/finetune_lora_single_gpu.sh>ValueError: Target module QuantLinear() is not supported.
#960
LIXUEGUANG002
closed
8 months ago
2
Qwen-14B-chat微调分布式微调无法启用<title>
#959
SuperTom99
closed
8 months ago
4
[BUG] 不遵循指令,输出和上一条重复
#958
huajinghua
closed
6 months ago
1
[BUG] run_gptq.py 对Lora微调并合并后的模型进行GPTQ4量化,直接把内存耗尽了
#957
skyantao
closed
5 months ago
2
messages中"role": "system"的顺序问题(上一个提问不小心关掉了)
#956
huangyunxin
closed
8 months ago
3
messages中"role": "system"的顺序问题
#955
huangyunxin
closed
8 months ago
0
zero2 lora微调内存异常升高,等待时间过长
#954
liyang1023
closed
8 months ago
3
撤销
#953
liyang1023
closed
8 months ago
0
fix bug: torchrun: error: argument --master-addr/--master_addr: expec…
#952
caiyueliang
closed
8 months ago
1
[BUG] <运行测试样例的时候机器重启>
#951
WyHy
closed
5 months ago
2
💡 [REQUEST] -Qwen-72B-Chat使用AutoGptQ量化完后的目录,跟官网公布的Qwen-72B-Chat-Int4目录存在一些配置文件差异
#950
lifengyu2005
closed
8 months ago
2
[BUG]AssertionError: libcuda.so cannot found!
#949
ArlanCooper
closed
8 months ago
9
基础加减乘除4则运算不能完全
#948
LiXuanming
closed
6 months ago
1
[BUG] kv cache报错
#947
sunchao2828
closed
6 months ago
1
[BUG] <单卡lora微调>
#946
yoohhwz
closed
5 months ago
8
进行微调,进度条始终没有动过,而单机多卡微调也出现奇怪的错误<title>
#945
SuperTom99
closed
5 months ago
2
vLLM和LORA的问题[BUG] <title>
#944
xx-Jiangwen
closed
6 months ago
2
Qlora 合并💡 [REQUEST] - <title>
#943
olin-Ai
closed
6 months ago
1
在readme中长文本理解中:“通过arXiv数据集上的语言模型实验,发现 Qwen 在长上下文场景下可以达到出色的性能”,这个arXiv数据集是怎么构造的?
#942
lvjianxin
closed
8 months ago
1
LoRA微调显存占用很大,A10*4
#941
Matteyes
closed
8 months ago
5
[BUG] Lora训练出来没有adapter_model.bin
#940
tianyouyangying
closed
8 months ago
1
[BUG] <单卡微调报错>
#939
yoohhwz
closed
8 months ago
1
[BUG] Input data may contain inappropriate content. api error...
#938
jun0wanan
closed
8 months ago
4
💡 [REQUEST] - <title>如何在Qwen14B和72B中应用SelfExtend
#937
ArcherShirou
closed
5 months ago
3
[BUG] <title> 使用 finetune_lora_ds.sh 脚本跑qwen-14B 的一级多卡lora精调,任务失败
#936
ghost
closed
8 months ago
1
[BUG] <title>微调完了之后,chat回答的不是微调的数据
#935
kriends
closed
5 months ago
6
Fix bug of fschat version in Dockerfile-cu121.
#934
fyabc
closed
8 months ago
0
训练代码中的modeling_qwen.py 中attention_mask.masked_fill的使用有问题导致 attention_mask不是下三角矩阵
#933
jru001
closed
8 months ago
1
💡 [REQUEST] - <title>多机多卡推理inference using multiple machines and multiple GPUs
#932
crowdin001
closed
6 months ago
2
[BUG] 千问1.8b在gpu推理且温度小于等于0.5情况下出错
#931
pengbj
closed
8 months ago
1
[BUG] Qwen-14B-Chat-Int4 GPTQ model is slower than original model Qwen-14B-Chat greatly
#930
micronetboy
closed
8 months ago
2
[BUG] Qwen-14B-Chat-Int4 GPTQ model using vLLM. Got this warning: Using a slow tokenizer. This might cause a significant slowdown. Consider using a fast tokenizer instead.
#929
micronetboy
closed
9 months ago
3
[BUG] RuntimeError: 'weight' must be 2-D
#928
lyc202001
closed
8 months ago
1
qwen-14b-chat-int4多卡推理,显存分配不均匀
#927
1424153694
closed
8 months ago
9
对70B base model和14B base model进行evaluate 72B loss的值 大于 14B loss
#926
wccccp
closed
5 months ago
5
[BUG] Qwen-14B-Chat-Int4 使用 vLLM 并没有加速?
#925
micronetboy
closed
8 months ago
4
cache_autogptq_cuda_356.cpp与cache_autogptq_cuda_kernel_245.cu有下载链接吗,内外网都搜了只找到256和256的
#924
zzwtop1
closed
8 months ago
3
[BUG] <title>
#923
zzwtop1
closed
9 months ago
0
Previous
Next