issues
search
lich99
/
ChatGLM-finetune-LoRA
Code for fintune ChatGLM-6b using low-rank adaptation (LoRA)
Apache License 2.0
726
stars
64
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
CUDA error: device-side assert triggered
#52
488283943
opened
8 months ago
0
报错
#49
zjbian
opened
1 year ago
0
报错
#48
fxb392
opened
1 year ago
0
不考虑GLM的双向注意力部分，注意力矩阵不是一个下三角矩阵吗
#47
fxb392
opened
1 year ago
0
显存问题
#46
zhangsanjava
opened
1 year ago
0
位置编码
#45
fxb392
closed
1 year ago
3
作者你好，这个项目对python的版本有什么要求吗，3.7是否可以？
#44
tjulh
opened
1 year ago
1
while cnt < retry_cnt:
#43
fxb392
opened
1 year ago
0
问答数据集如何构建
#42
Godlikemandyy
opened
1 year ago
0
请问有交流群吗?
#41
roki1031
opened
1 year ago
0
有大佬们试过用8卡训练的吗？
#40
tu2022
closed
1 year ago
0
请问LORA模型参数怎么加载到原模型里呢？
#39
ZeyuBa
closed
1 year ago
0
peft 0.3.0如何设置adapater_name
#38
moseshu
opened
1 year ago
8
context_length = obj['prompt'].index(130004)
#37
moseshu
closed
1 year ago
2
LoRA训练时间大概是多久呢？
#36
realcarlos
opened
1 year ago
3
finetune没效果
#34
ChenBinfighting1
opened
1 year ago
4
关于ZeRO的疑问？
#33
MAxx8371
opened
1 year ago
3
example.ipynb中进行训练测试loss为nan
#32
SilentMoebuta
opened
1 year ago
1
Fix some minor typos
#31
l0rinc
closed
1 year ago
0
[deepspeed] OVERFLOW!
#30
JingerAI
opened
1 year ago
1
No such file or directory: '/root/.cache/huggingface/modules/transformers_modules/chatglm-6b/tokenization_chatglm.py'
#29
Data2Me
opened
1 year ago
8
ValueError: 150004 is not in list是什么回事？
#28
z1968357787
closed
1 year ago
3
torch.distributed.elastic.multiprocessing.errors.ChildFailedError, when running the train_new.py
#27
Skywalker-Harrison
opened
1 year ago
2
这个显卡要求一定是bfloat16吗
#26
z1968357787
opened
1 year ago
3
关于分布式GPU训练
#25
z1968357787
opened
1 year ago
1
example_simple报错
#24
qishisurenhhh
opened
1 year ago
1
here are my questions,I have more than 4 gpus to run the train.py,but it still out of memory,I check the usage of memory and find that one of them overflows and produce the bug,how can I solve it?
#23
z1968357787
opened
1 year ago
0
LoRA的A矩阵一直不更新
#22
qz701731tby
opened
1 year ago
1
您好，问一下，这个训练完有可以展示与原来基础模型进行对比的测试效果吗
#21
kunshou123
opened
1 year ago
0
单机多卡报错
#20
ForgetThatNight
opened
1 year ago
0
训练loss变为NaN
#19
qz701731tby
closed
1 year ago
2
已获取
#18
lbxcfx
closed
1 year ago
0
subprocess.CalledProcessError: Command '['ninja', '-v']' returned non-zero exit status 1.
#17
xiamaozi11
opened
1 year ago
2
RuntimeError: CUDA error: invalid device ordinal
#16
yanqiangmiffy
closed
1 year ago
1
有什么卡训练的，V100完全搞不定
#15
huzaizi2023
opened
1 year ago
11
LORAConfig报错：ValueError: Target modules ['q', 'k', 'v'] not found in the base model. Please check the target modules and try again.
#14
nameless0704
closed
1 year ago
6
train.py的命令行启动是什么？
#13
Data2Me
closed
1 year ago
1
如何设置batch_size个数，变动后train会变维度上的错误
#12
GUORUIWANG
closed
1 year ago
1
可以给一下具体的环境requirement吗?
#11
Data2Me
closed
1 year ago
1
请问DeepSpeed与Multi-gpu是绑定的吗？使用DeepSpeed提升效果有多大？
#10
nameless0704
closed
1 year ago
2
About multi-GPU
#9
zhongtao93
closed
1 year ago
6
请问下Finetune之后能实现企业定制FAQ的效果吗？可能有一百个问答这样
#8
terryops
opened
1 year ago
8
能否使用量化后的chatGLM-6b-int4小模型进行微调？
#7
valkryhx
closed
1 year ago
4
两个epochs之间，loss值并没有降下来
#6
aizpy
closed
1 year ago
1
NameError: name 'train_dataloader' is not defined
#5
wccccp
closed
1 year ago
1
训练后的结果对应不上
#4
zhangyanbo2007
closed
1 year ago
1
训练超显存
#3
GaoPengGit
closed
1 year ago
5
数据集和微调模型的一些问题
#2
SarmonFish
closed
1 year ago
1
Is there any details about dataset?
#1
980202006
closed
1 year ago
2