Skip to content

Issues: Tencent/TencentPretrain

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.

Sign up for GitHub

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

31 Open 5 Closed
31 Open 5 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

SMP2020-EWECT的数据集开源使用问题
#133 opened Oct 31, 2024 by pxaklbe
3
请问这里的中文模型支持的最大输入序列长度是512tokens吗？超过512tokens就会被截断嘛？可不可以在微调的时候扩大模型的位置编码数量？
#130 opened Jul 21, 2024 by chengzi-big
单机2卡预训练LLAMA-7B报错TypeError: an integer is required (got type NoneType)
#112 opened Nov 29, 2023 by smallYellowCat
1
【问题】deepspeed如何对不同显存大小分配数据，我有32G和16G两种大小的GPU
#98 opened Sep 26, 2023 by 18liumin
1
LLaMA2-70B格式转换
#96 opened Sep 5, 2023 by Double-bear
size mismatch for classifier.weight: copying a param with shape torch.Size([7, 768]) from checkpoint, the shape in current model is torch.Size([2, 768]).
#93 opened Aug 19, 2023 by beat4ocean
KeyError: 'd'
#92 opened Aug 19, 2023 by beat4ocean
1
Resume from checkpoint
#81 opened Aug 15, 2023 by mohammadaminabbasi
发现几个bug，dynamic_masking 意思是不是写反了，多了not
#80 opened Aug 10, 2023 by boluoyu
lora推理
#71 opened Jun 6, 2023 by aSmallsheep
1
DeepSpeedZeRoOffload initialization failed (can't allocate memory)
#70 opened May 25, 2023 by treya-lin
关于扩充词表再增量预训练的疑问
#68 opened May 19, 2023 by qiancheng99
关于LLAMA预训练的疑问
#66 opened May 17, 2023 by zhanghaok
1
有计划支持torch2.0的flashattention这些吗
#61 opened May 11, 2023 by zhuojianc
训练完怎么执行推理，发现训练保存的模型只有17M
#59 opened May 4, 2023 by jiangjingyao
1
似乎没有保存rotary_emb.inv_freq的权重
#58 opened May 2, 2023 by cristianoc20
10
pretrain.py of llama-7b model, Exception: : Current loss scale already at minimum
#57 opened Apr 26, 2023 by liukaiyueyuo
2
单机8卡A100-80G deepspeed ZERO3 或者 非ZERO3 pretrain LLaMA-7B时，不能充分利用显卡
#56 opened Apr 25, 2023 by ShadowTeamCN
1
支持bloom模型的预训练吗
#53 opened Apr 18, 2023 by nishiwen1214
llama训练时，best状态存储导致训练卡顿，建议删除存储best文件部分代码，望记得更新
#52 opened Apr 16, 2023 by baketbek
4
lora训练llama 貌似不支持？
#51 opened Apr 15, 2023 by wind91725
3
Any plan of supporting Bloom model?
#50 opened Apr 15, 2023 by wind91725
多卡跑llama预训练出错：RuntimeError: CUDA error: invalid resource handle，Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.
#43 opened Apr 11, 2023 by baketbek
1
lora貌似不支持ddp？
#41 opened Apr 11, 2023 by wutaiqiang
用自己的中文数据的话，preprocess中需要把数据格式调整成什么形式即可？这部分相关说明有吗？目标是想做llama的增量预训练
#40 opened Apr 10, 2023 by baketbek
1
ProTip! no:milestone will show everything without a milestone.