issues
search
THUDM
/
ChatGLM2-6B
ChatGLM2-6B: An Open Bilingual Chat LLM | 开源双语对话语言模型
Other
15.73k
stars
1.85k
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
[Feature] 麻烦自己怎么离线量化呢？有方法吗
#628
lonngxiang
opened
1 year ago
0
[Help] 请问api.py中的max_length参数具体指的是什么length
#627
zhengyangyong
opened
1 year ago
0
[Help] 模型在推理时是否存在记忆？
#626
CoderHiro
closed
1 year ago
0
UnicodeDecodeError: 'utf-8' codec can't decode byte 0xce in position 5811: invalid continuation byte
#625
123eye
opened
1 year ago
0
[单卡支持多个Lora] <如何在微调的时候支持多个Lora 同时训练>
#624
merlintang
opened
1 year ago
0
[BUG/Help] ptuning微调之后向模型提问，返回空
#623
2021QKA
opened
1 year ago
5
[Help] API部署的情况下，怎么打断模型执行？终止生成
#622
mymynew
opened
1 year ago
0
[BUG/Help] 单张A800 显存80g 跑chatglm2没问题，但是使用两张A40，一张A40显存48g，跑chatglm2报了torch.cuda.OutOfMemoryError: CUDA out of memory.
#621
zhengdacheng
opened
1 year ago
1
[BUG/Help] 在ptuning中的main.py中，第130行model = model.half()，采用cpu加载，造成超内存报错
#619
zhuzongjian1
opened
1 year ago
1
[新增低资源语言] <如何新增加一个新的语言>
#618
huqitu
opened
1 year ago
0
[Help]关于source_prefix参数的疑问
#617
YinHan-Zhang
opened
1 year ago
0
[BUG/Help] <title>运行web_demo.py报错
#616
gotcha001x
opened
1 year ago
3
[Help] <ptuning时，增大target_length后训练，模型不收敛>
#614
qdchenxiaoyan
opened
1 year ago
1
[BUG/Help] model = AutoModel.from_pretrained("D:\\ChatGLM\\model\\2", trust_remote_code=True).cuda() 没有报错直接退出
#613
zhans1099
opened
1 year ago
3
[BUG/Help] <window环境下微调模型>
#612
hfut-GYH
opened
1 year ago
0
[BUG/Help] Readme中进行 openapi-api的API 调用的示例代码报错
#611
yecllsl
opened
1 year ago
2
[BUG/Help] <ptuning的微调方式，合并权重到basemodel的代码和方法有么？>
#610
cyrishe
opened
1 year ago
0
[BUG/Help] <title>请问是否支持设置beam size >1，并将生成的多个输出一起返回呢？
#609
kuranatsu
opened
1 year ago
0
[BUG/Help] 请求大神对微调参数设置进行详解
#608
YueleiFu
opened
1 year ago
3
[BUG/Help] 使用 adapter 方法微调模型在 forward pass 出现 nan / inf 异常值
#606
ZionDoki
opened
1 year ago
0
[BUG/Help] <title>如何提高多卡下的p-tuning训练速度
#605
zhengw1024
opened
1 year ago
1
[BUG/Help] 到底是我不懂中文还是ChatGLM不懂英文，总线=Bus=公交车
#604
trouvaille2023
opened
1 year ago
1
left padding side和mask下三角矩阵之间的矛盾
#603
nostalgiaer
opened
1 year ago
0
RuntimeError: CUDA error: an illegal memory access was encountered
#602
leoluopy
opened
1 year ago
1
[BUG/Help] <title>
#601
Hubotcoder
opened
1 year ago
0
如何改造原有模型并达到私有化使用。
#600
TzyTman
opened
1 year ago
3
有打算开源比6B更强的模型吗
#599
zengzikang
opened
1 year ago
0
[BUG/Help] <为什么同一个字符会对应多个token_id？（排除BPE分词处理）>
#598
ito-integration
opened
1 year ago
0
逻辑能力有待加强
#597
315930399
opened
1 year ago
0
运行web_demo.py，向GLM提问遇到JS错误
#596
wrl1224
opened
1 year ago
4
[BUG/Help] <title>ChatGLM可以在一个服务中同时加载多个lora微调模型，根据输入切换使用哪个模型吗？
#595
lrx1213
opened
1 year ago
0
[Help] 长文本推理OOM
#594
Wohoholo
opened
1 year ago
3
【Help】使用lm-evaluation-harness评估，ChatGLM2-6B在CEval上准确率很低？
#593
Kevin-KWH
opened
1 year ago
2
执行train.sh 进行微调报错
#592
TzyTman
opened
1 year ago
0
为什么提问是调用的https://api.segment.io/v1/t这个url，必须是这样吗还是可以配置
#591
markyangm
opened
1 year ago
0
[BUG/Help] RuntimeError: CUDA error: CUBLAS_STATUS_NOT_INITIALIZED when calling `cublasCreate(handle)`
#590
Alan-JW
opened
1 year ago
1
[BUG/Help] 基于文档问答的stream输出和非stream输出不一致，stream输出会有格式混乱的情况
#589
y111x
closed
1 year ago
0
[BUG/Help] 6b-int4 模型全量微调 loss 为 nan
#588
wizardforcel
closed
1 year ago
0
chatglm2微调回合数该如何设置[BUG/Help] <title>
#587
renllll
opened
1 year ago
1
[BUG/Help] openai_api.py调用报错404
#586
Tanorika
opened
1 year ago
1
模型部署依赖哪些包，为什么在不同平台部署性能不一致
#585
Connor-Shen
closed
1 year ago
0
[Help] lora微调合并后模型推理速度明显慢了好多
#583
daydayup-zyn
opened
1 year ago
2
想问问数据集中的标签有什么特殊含义吗？可以随意修改吗？
#582
sunyrain
opened
1 year ago
1
请问是否支持在国产dcu上部署呢？[Feature] <title>
#581
JACKzhuz
opened
1 year ago
1
[Help] <title请问一下，如果训练可以让chatglm2-6b获取时间
#579
lianyifeng
opened
1 year ago
0
多种prompt语料类型同时训练会影响准确度？
#578
RainBowHEric
opened
1 year ago
0
[BUG/Help] <title>流式API部署
#577
Hannieliao
opened
1 year ago
1
微调
#576
ysq873
opened
1 year ago
0
[Help] 训练的时候NOTE: Redirects are currently not supported in Windows or MacOs.
#575
Zona007
opened
1 year ago
0
[BUG/Help] <title>本地调用模型失败，报错'ChatGLMTokenizer' object has no attribute 'tokenizer'
#574
arx-night
opened
1 year ago
5
Previous
Next