-
Notifications
You must be signed in to change notification settings - Fork 422
Issues: shibing624/MedicalGPT
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Author
Label
Projects
Milestones
Assignee
Sort
Issues list
运行pretraining.py时报错:RuntimeError: CUDA error: device-side assert triggered
bug
Something isn't working
#376
opened May 16, 2024 by
Wenting1227
ppo训练时出现问题:UserWarning: KL divergence is starting to become negative: -233.50
question
Further information is requested
#374
opened May 15, 2024 by
user2311717757
Regarding RLHF and DPO training data
question
Further information is requested
#358
opened Apr 3, 2024 by
Aniketto16
使用deepspeed 全参数sft后,inference 回答的都为空,有解决办法吗
question
Further information is requested
#357
opened Apr 2, 2024 by
Yian320
ValueError: operands could not be broadcast together with remapped shapes [original->remapped]: (3,2) and requested shape (1,2)
bug
Something isn't working
#356
opened Mar 27, 2024 by
Riapy
扩充词表后能否直接进行SFT呢?
question
Further information is requested
#352
opened Mar 24, 2024 by
HaotianLiu123
预训练后模型出现自问自答、输出未知序列、重复口吃现象
question
Further information is requested
#351
opened Mar 21, 2024 by
Peter-of-Astora
llama进行rm训练的时候,出现问题ValueError: weight is on the meta device, we need a Something isn't working
value
to put in on cpu.
bug
#347
opened Mar 14, 2024 by
cove1011
使用qwen进行pretrain的时候出现了问题:Cannot copy out of meta tensor; no data!
bug
Something isn't working
#346
opened Mar 12, 2024 by
cove1011
单机多卡sft deepspeed zero3 训练一直卡在训练阶段
question
Further information is requested
#330
opened Feb 12, 2024 by
lainxx
请问,pt阶段,基础模型比较大(Yi-67B),多机多卡用那种训练比较好呢?
question
Further information is requested
#315
opened Jan 23, 2024 by
listwebit
在单机多卡监督微调时使用的策略是DP还是DDP?
question
Further information is requested
#291
opened Dec 18, 2023 by
CNUIGB
请问大佬,Reward model验证分类评分,一个问题回传两个tensor?
question
Further information is requested
#284
opened Dec 11, 2023 by
waycup7
大佬,使用自己数据进行增量预训练时,loss不降反增。
question
Further information is requested
#280
opened Dec 5, 2023 by
SevenMpp
Previous Next
ProTip!
Find all open issues with in progress development work with linked:pr.